Abstract
The use of computational fluid dynamics (CFD) in the wind engineering (WE) is generally defined as computational wind engineering (CWE). Since its foundation in 2004, the use of OpenFOAM in CWE has been increasing progressively and covers nowadays a wide range of topics, from wind environment to wind structural engineering. This paper was drafted in response to the invitation from the organizers of the 18th OpenFOAM workshop held in Genoa (Italy) on 11–14 July 2023, when a technical session on Civil Engineering and Wind Engineering was organized. In this paper the author briefly reviews the history of WE and surveys the evolution, methods, and future challenges of OpenFOAM in the CWE. Topics are here regrouped into three main research areas and discussed from a physical, engineering and purely computational perspective. The study does not cover the Wind Energy and related topics, since this can be considered nowadays as a stand-alone subfield of the WE. This review confirms that OpenFOAM is a versatile tool widely used for WE applications that often require new models to be developed ad hoc by CFD users. It can be coupled easily with numerical weather prediction models for mesoscale-microscale wind and thermal studies, with building energy simulation models to determine the energy demand, with finite element method for structural engineering design. OpenFOAM represents an extraordinary opportunity for all CFD users worldwide to share codes and case studies, to explore the potential of new functionalities and strengthen the network within the CFD community.
Similar content being viewed by others
Explore related subjects
Find the latest articles, discoveries, and news in related topics.Avoid common mistakes on your manuscript.
1 Introduction
In the last decades and since its birth, several definitions of wind engineering (WE) have been provided by scholars and researchers. Cermak [1] best defined the WE as the “rational treatment of interactions between wind in the atmospheric boundary layer and man and his works on the surface of Earth”. A broad and fascinating definition has been recently provided by Solari [2], who stressed out the dualism property of the wind. Solari [2] defined the wind as the evil when it destroys structures and infrastructures, makes environments uncomfortable, erodes lands by leading to desertification, causes fatalities. On the other hand, the wind may be good because it is the engine of atmospheric motions, it allows the production of green renewable energy through windmills and wind turbines, it blows far away pollutants from cities, and it improves lives of people in deserts and arid lands through natural ventilation. As reviewed by Solari [3], the WE discipline has been evolving and adapting itself through the centuries, by progressively involving more and more topics in which the wind plays a major role. This is also evident from the evolution of session topics scheduled in the WE conferences through the last decades. In 1963, in what we might consider the seed of the International Wind Engineering Conference (ICWE), namely the Symposium on Wind Effects on Buildings and Structures held in Teddington, UK, contributions mainly devoted to wind effects on structures were presented [4, 5]. In 2023, in the most recent 16th International Wind Engineering Conference held in Florence, Italy, a list of thirty-one potential topics spanning from meteorological phenomena to pollution dispersion was presented [6].
Various experimental and numerical techniques have been supporting the WE research studies through different eras (e.g. [7,8,9]). In most of the cases, these techniques have been inherited from other communities (as aerospace and mechanical engineering) and further developed to address WE topics. By overlooking the experimental techniques (as wind tunnels and field measurements) not covered by this review, the numerical technique mostly used by the WE community is certainly the computational fluid dynamics (CFD). As mentioned by Blocken [10], the CFD made its firstly appearance in a WE conference as topic session in 1991, at the 8th International Wind Engineering Conference held in London, Ontario, Canada, and chaired by A.G. Davenport [11]. The use of CFD in WE is generally defined as computational wind engineering (CWE). The scientific literature offers a long list of outstanding review papers focusing on the history and evolution of the CWE (e.g. [8, 12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27]). The 1st International Symposium on Computational Wind Engineering was held at the University of Tokyo in August 1992, and organized by Prof. S. Murakami also considered as the father of this discipline. In 1997 after 5 years the 1st CWE conference, Prof. Murakami reviewed the progress of CWE focusing on the mainly used numerical approaches, as the steady and unsteady Reynolds-averaged Navier–Stokes (RANS) and large-eddy simulation (LES) [13]. He concluded that although various numerical approaches were used to predict the wind flow around bluff bodies, how to best select such approaches (based on application) was still an open question at that time. He pointed out that albeit LES already gave better results than RANS, the former was too computationally demanding for many applications. Murakami [13] closed the paper by claiming that “… rapid evolution of CPU hardware will surely overcome this restriction, and wide application of LES to CWE problems will certainly be realized in the near future”. It is curious to observe that, despite the rapid growth in computing power of the last two decades, the use of LES in several CWE topics still remains prohibitive for academicians as well as for practitioners. On the other hand, less sophisticated (then less computationally demanding) approaches like the steady RANS became the most used in CWE research and practice through the years (e.g. [7, 10, 28,29,30]). Meantime, extensive best practice guidelines documents have been developed to help the CFD users with the setting of approximate form of governing equations, turbulence model, domain and grid, boundary conditions, discretization schemes, and so on (e.g. [31,32,33,34,35]). Likewise, as reviewed by Blocken [10] the development of verification and validation as well as of uncertainty quantification methods in CFD for WE simulations became crucial to support the activities of both scientists and practitioners (e.g. [36,37,38,39]). On this subject, an exhaustive review has been recently provided by Tominaga et al. [40]. It is worth noticing that if few years ago the mostly used approach (i.e. RANS) was strongly supported by best practice guidelines, nowadays similar guidelines are almost totally missing for one of the most currently used approaches (i.e. the LES). As detailed by Potsis et al. [27], some standardized procedures have been provided about pressure distributions and peak structural response on building (e.g. [24, 40]). However, except for some stand-alone contributions, there is no consensus in the scientific community about the verification and validation process. The state-of-the-art lacks generalized LES best practice guidelines that may help the user in defining (as for RANS) the most relevant computational settings: domain size, grid size (in LES responsible for the cut-off filtering), the choice of the most appropriate inflow method generator (rather relevant in LES), and so on. Moreover, also the validation process (with experimental data) of LES results may be more intricate than RANS results. LES data should rely not only on mean quantities but also on high-order statistics in order to accurately describe the turbulent structures of the flow. Overall, these aspects can be considered as serious concerns that sometimes may limit the applicability of LES to a wide range of numerical studies [30].
The CFD tools historically used in CWE are mainly two: the commercial software ANSYS Fluent [41] and the open-source software OpenFOAM [42, 43]. However, to the best knowledge of the author, a fully dedicated review on scientific CWE studies carried out with OpenFOAM is still missing. This is the innovative scope of the present paper in which CWE topics investigated with OpenFOAM in the last decades are reviewed. Since the 2000s, OpenFOAM has been widely used by wind engineers as well as atmospheric scientists to simulate the wind and airflow. OpenFOAM as well as other open-source codes has some advantages with respect to commercial CFD software (e.g. ANSYS Fluent but not only): (1) the user is not bound to pay any license; (2) codes from the pre-processing to post-processing stage can be manipulated and implemented from scratch; (3) it can be integrated with other open-source codes and vice-versa (e.g. Phyton, Paraview, WRF, etc.); (4) since free-license the user can run simulations in parallel on high-performance computing (HPC) systems with a large number of nodes (i.e. cores) without worrying about license costs. On the other hand, OpenFOAM requires programming skills (e.g. C++) with respect to other commercial tools, making this software not accessible at first glance to users that may lack the required know-how in programming. Moreover, OpenFOAM is not based on a graphical user interface (GUI) and this could make the software generally less intuitive than other commercial CFD tools.
This paper was drafted in response to the invitation from the organizers of the 18th OpenFOAM workshop held in Genoa (Italy) on 11–14 July 2023, for which technical sessions on Civil Engineering and Wind Engineering were organized. For the present review, a database of 289 papers, books and official websites has been assembled. The papers have been selected from well-known website platforms (as Science Direct) by using the most common keywords appeared in the international WE and CWE conferences of the past decades: wind, flow, CFD, OpenFOAM, building, bridge, complex terrain, downscale, mesoscale, microscale, microclimate, wind comfort, wind-driven rain, wind load, windblown sand, snow drift/accumulation, and so on. As expected, most of the papers (about 34%) have been published on the reference journal of the WE and CWE community, that is Journal of Wind Engineering and Industrial Aerodynamics, followed by Computers and Fluids, Building and Environments, Urban Climate, Applied Mathematical Modelling, Boundary-Layer Meteorology, and so on. An overview about the use of OpenFOAM in the CWE is provided in Fig. 1. Three main research areas have been identified in this review: (1) CFD for complex terrain: from mesoscale to microscale, (2) CFD for the built environment, (3) CFD for wind structural engineering. Within these research areas, various topics have been selected in turn. The statistics related to the use of OpenFOAM per research area and topic are reported in Fig. 1a and b, respectively. The percentage indicates the number of papers reviewed per research area (Fig. 1a) and per topic (Fig. 1b), with respect the total number of OpenFOAM studies in CWE considered in this review, namely 188. The histogram of Fig. 1c shows the OpenFOAM studies in CWE published from 2011 to 2023. Some papers have been also published earlier, between 2004 (i.e. the foundation of OpenFOAM) and 2011, but these are considered statistically not meaningful for the final purpose of this plot. Note that the Wind Energy is one of the topics where OpenFOAM is most widely used. However, the Wind Energy can nowadays be considered as a stand-alone WE subfield, with its own community and international conferences worldwide. The same applies also for topics closely related to Wind Energy, dealing with wind and wave effects on off-shore structures (as off-shore/on-shore wind turbines, wave–structure interaction, floating structures, and so on). For this reason, only research studies closely related to wind energy exploitation in the built environment are included in the present review.
The geometry and grid generation is a crucial stage to achieve successful CFD results, irrespective of the CWE study under investigation. For sake of brevity, in the present paper the author decided not to focus on very detailed (although important) geometry and grid generation procedures nor on the tool used to generate them. Nevertheless, it is worth noticing that the studies carried out with OpenFOAM and mentioned here do not always use grids constructed with open-source tools/utilities of OpenFOAM (as SnappyHexMesh) or integrated in it (as CfMesh). Conversely, in most of the cases either other open-source tools or commercial tools have been used to construct geometry and grid.
The paper is organized as follows. Section 2 discusses the use of OpenFOAM for complex terrains, from mesoscale to microscale. Section 3 discusses the use of OpenFOAM for the built environment, in particular, for wind and thermal comfort assessment, pollutant/gas dispersion, wind-driven rain, windblown sand and snowdrift. Section 4 addresses the use of OpenFOAM for the modeling of wind effects/loads on bridges, low-rise and high-rise structures as well as photovoltaic panels. Section 5 closes the paper with summary and perspectives.
2 CFD for complex terrains: from mesoscale to microscale
This section focuses on the use of OpenFOAM for the simulation of winds on complex terrains, from the mesoscale to the microscale. The contributions on this topic represent approximately the 12.23% of the total number of OpenFOAM contributions of this review (see Fig. 1a).
When the micro-scale (see [44] for the classification of meteorological scales) model OpenFOAM is applied to simulate real full-scale environmental flows, like wind flows over complex terrain as well as in urban areas, some information concerning the larger-scale (i.e. meso-scale) meteorological conditions is required. This can be obtained either using the micro-scale model in standalone configuration, implementing suitable initial and boundary conditions by means of analytical models or interpolation of the needed variables (e.g., wind velocity, turbulent kinetic energy, energy dissipation rate, temperature, etc.), or coupling OpenFOAM to a meteorological model (or similar proxy like reanalysis) through a process of downscaling of meteorological fields from the larger-scale model to the micro-scale one. In the micro-scale model, the complexity of terrains is reproduced numerically as combination of two stages. The STereo Lithography interface format (STL) is usually used to accurately reproduce the complex orographic condition. The non-homogeneous roughness is generally modeled by means of patches necessary to subdivide the ground in terms of homogeneous aerodynamic roughness length (\({z}_{0}\)). Additionally, when the orography is very complex and steep (e.g. mountains or hills), it is common practice to add a buffer zone with artificially smoothed orography at the external boundaries to avoid artificial gradients responsible for potential numerical instabilities (e.g. [45, 46]). Urban texture can also be included in the STL file to explicitly model structures and infrastructures. Once the three-dimensional geometry, the computational domain and grid are generated, the minimal approach to run standalone CFD simulations consists in assuming a neutral atmosphere, i.e. thermal energy equation is neglected. In such a case, standard inlet conditions are defined in terms of vertical profiles (e.g. power law, log-law-of-the-wall, and turbulence [34, 47, 48]) for steady/unsteady RANS simulations [49], while inflow generation methods must be adopted for LES simulations. Inflow generation methods have been widely applied in OpenFOAM, as the method of Deodatis [50] used by Ren et al. [45] to simulate wind fields in complex orography and more recently the one proposed by Melaku and Bitsuamlak [51] validated by means of wind-tunnel tests.
Simulating non-neutral atmospheric conditions in standalone micro-scale models is more challenging. The wind field simulation using RANS/URANS requires inlet wind profiles following the Monin–Obukhov similarity as well as modified turbulence model and wall functions to avoid turbulence decay within the computational domain [52]. The consideration of thermal boundary conditions can also require ad-hoc implementations to make the temperature inlet profile consistent with wall functions [53]. For LES applications, inflow generation methods like the method proposed by Xie and Castro [54] exist, these are suitable for neutral as well as unstable atmospheric conditions and were tested by Vonlanthen et al. [55] over a real terrain topography.
The dynamic downscaling approach refers to the use of meso-scale models to explicitly resolve the effects of the larger-scale meteorological phenomena coupled dynamically to micro-scale models to capture the local phenomena of main interest for the CWE. Meso-scale models are usually meteorological models used to forecast weather conditions, also commonly known as numerical weather prediction (NWP) models. In principle, the coupling between NWP models and micro-scale CFD models can be realized through a one-way or two-way technique. For the one-way coupling of NWP models with OpenFOAM steady RANS accounting for thermal effects (i.e. buoyantBoussinesqSimpleFOAM solver), the larger-scale model (the NWP) is used to calculate the steady state of the atmosphere at each output time step, which is typically in the order of few hours. This information is used in turn to evaluate the boundary conditions at different times for the micro-scale model (the CFD) in terms of wind and turbulence profiles using standard analytical function (e.g. [47]) or simply interpolating from the coarse to the fine grid. Accordingly, this approach needs proper post-processing algorithms to extract the required information (e.g. wind, temperature, etc.) from the NWP model and allows running the micro-scale model independently from the larger-scale one, which in theory makes also straightforward the coupling of micro-scale models with available datasets of hindcast or reanalysis. No feedback is expected in one-way coupling from the CFD model to the NWP model. Miao et al. [56], for instance, coupled the Weather Research and Forecast (WRF) by Skamarock et al. [57] to OpenFOAM in order to simulate the pollutant dispersion in Beijing, China. The resolution of the WRF inner domain was of 1 km, equal to the micro-scale domain size, so that only the grid cell nearest to the CFD inlet was used to provide the boundary conditions for CFD simulations. Analogous studies were carried out by various scientists: Wong et al. [58] inestigated the urban microclimate in Singapore; Kadaverugu et al. [59] simulated the wind fields in Nagpur City (India) accounting for the vegetation effects with the porousSimpleFOAM solver; Castorrini et al. [60] and Che et al. [61] applied the one-way coupling approach for wind resource prediction purposes. A systematic analysis of the one-way coupling WRF-OpenFOAM for different turbulence models available in WRF and in OpenFOAM is reported by Temel and van Beeck [62]. By testing various turbulence models for the Askervein Hill project, the authors proposed several closure constants and near-wall treatment suitable for near-neutral conditions [63]. Following up this work, the authors presented a modified version of the PBL BouLac scheme for WRF coupled to OpenFOAM that improves the prediction of velocity and turbulence by using the RANS approach in complex terrain [64]. The authors further refined closure constants and near-wall treatment [65] based on simulations of the Joint Urban 2003 experiment [66]. Based on the OpenFOAM micro-scale RANS simulations of the Askervein and the Joint Urban 2003 field campaigns, García-Sánchez and Gorlé [67] tried to evaluate the wind field uncertainty of the one-way coupling technique with an ensemble approach. Similarly to the ensemble prediction methods used in weather forecast, in the ensemble approach the CFD boundary conditions (calculated from the NWP model) are perturbated and the variability of the CFD results properly quantified. They concluded that the ensemble one-way coupling can be considered satisfactory since it encompasses with high confidence level the true solution.
In the one-way coupling of NWP models with OpenFOAM in URANS or LES cases (i.e. buoyantBoussinesqPimpleFOAM solver), the larger-scale model provides the CFD model with (fixed) initial and (time-varying) boundary conditions. Mughal et al. [68] coupled the WRF-LES model to OpenFOAM-LES using a post-processor to un-staggering the required WRF variables (i.e. velocity, pressure and temperature) on a collocated grid for each available time step (e.g., every 30 s) of the NWP model. Variables are then used as Dirichlet conditions at the OpenFOAM inlet through the timeVaryingMappedFixedValue boundary condition. Ground temperature and thermal fluxes are also assimilated from WRF. Similarly, Piroozmand et al. [69] coupled COSMO-URANS to OpenFOAM-URANS in order to simulate the wind and temperature fields in city of Zurich (Fig. 2). The two procedures are conceptually analogous, with the main exception that in URANS the time-step size can be larger with respect to LES with implications on the Courant number and inlet conditions may differ in the two approaches. Another research study with URANS-URANS one-way coupling of WRF-OpenFOAM was presented recently by Castorrini et al. [70]. The authors applied this approach for wind prediction offshore taking into account also the sea wave formation, which increase turbulence generation at the surface. In the two-way coupling, models exchange information back and forth among each other in such a way that the micro-scale model also influences the NWP forecast. On one hand, this approach may theoretically lead to more accurate results (both at the larger and smaller scales) with respect to the simpler one-way coupling. On the other hand, it is more difficult to implement this approach especially when NWP and OpenFOAM models are involved, since both have to run in parallel with synchronized time step by exchanging information at a fast rate. To the best knowledge of the author, this methodology has not yet been explored in the scientific literature.
Overall, the high flexibility that OpenFOAM provides in a coupling process (with respect to other commercial CFD tools) is crucial to deeply investigate complex atmospheric phenomena at different spatiotemporal resolutions and favor the bridging between the Atmospheric Science and Wind Engineering communities.
3 CFD for the built environment
This section focuses on the use of OpenFOAM for simulation of wind in built environments, which represents approximately the 57.98% of the total number of OpenFOAM contributions of this review (see Fig. 1a). This section is organized into four subsections: wind flow modeling in the built environment (Sect. 3.1) with 19.68% of contributions, pedestrian wind comfort and urban microclimate (Sect. 3.2) with 12.23% of contributions, pollution and gas dispersion (Sect. 3.3) with 11.70% of contributions, wind-driven rain, windblown sand and snowdrift (Sect. 3.4) with 14.36% of contributions (see Fig. 1b).
3.1 Wind flow modeling in the built environment
Most of the best practice guidelines developed in the past two decades emphasized the importance of accurately modeling wind flow in built environment (e.g. [32, 34]). These recommendations have been used in a wide range of research studies, from the pedestrian-level wind comfort assessment to the exploitation of the wind energy by means of small-size wind turbines mounted on buildings’ roof (e.g. [71]). Best practice guidelines have been mostly developed with the aim of modeling by CFD (i.e. by RANS) neutrally stratified ABL winds in natural and built environments. According to Barlow [21] the Urban Boundary Layer (UBL)—the layer of the ABL where most of the population lives—consists of three main parts: the urban canopy layer (UCL), the layer up to the mean roof height where flow channeling occurs along the streets; the urban roughness sublayer (RSL) defined as the region including the UCL where the buildings exert significant drag on the flow and the wind is highly spatially dependent; and the inertial sublayer (IS) located above the RS where turbulence is homogenous and fluxes vary little with height (Fig. 3). The wind flow dynamics is generally quite complex in the UCL, where streets and open spaces (e.g. squares and green parks) may cause the onset of canyoning effects strongly dependent on the approach wind direction and urban fabrics [72]. As reported by previous review papers (e.g. [20, 73, 74]) a large number of studies have been carried out by scientists worldwide in order to investigate the wind flow behavior in the UBL. These studies were mainly targeted at providing new analytical models of vertical wind profiles (in terms of velocity and turbulent characteristics) in and above the UCL by using both experimental and numerical techniques (e.g. [75]). Most of these studies are generally carried out on arbitrary simplified geometries of buildings and cities. This is necessary in order to make the investigation less case specific and construct high-resolution computational geometries and grids that allow in turn to keep under control the number of control volumes. Probably one of the first attempts to simulate wind flow with OpenFOAM on a cluster of realistic buildings was made by Flores et al. [76]. The unsteady RANS and detached eddy simulation (DES) approaches were adopted to simulate the internal ventilation within an open pit including the effects of developed turbulence, buoyancy and stratification. A step forward was made by Carpentieri and Robins [77], who investigated by wind-tunnel tests and CFD simulations the impact of morphological parameters on wind flow in and above the UCL, through a simplified and a more realistic building array. The authors found that the building aspect ratio, the angle between street canyons and reference wind direction, and local geometrical features can significantly influence the flow dispersion. After a few years, a similar study aimed at understanding the impact of geometrical simplifications on wind flow in a realistic urban environment was published by Ricci et al. [78]. Here, results from 3D steady RANS and wind-tunnel tests were compared in terms of mean velocities and turbulent kinetic energy. The results showed that although simplification of geometries is generally possible and sometimes really necessary, more detailed CFD geometries can provide a better agreement with experiments (Fig. 4). Paden et al. [79] further investigated this aspect and proposed a workflow for automatically reconstructing simulation-ready 3D city models. As stated by the authors at the time of publication, the research study was still at an early stage and more work was required to further improve the reconstruction algorithms. Furtak-Cole and Ngan [80] proposed an alternative approach based on the vortex dynamics. This approach was applied to the UCL wind flow and a method for the prediction of mean horizontal velocities was derived. The Improved Delayed Detached Eddy Simulation (IDDES) turbulence model was then used to calibrate the approach on different built environment scenarios, for idealized and realistic canyons. The Delayed Detached Eddy Simulation (DDES) [81] as well as the IDDES that also incorporates the wall modeled LES (WMLES) [82] are among the most recent developments of the DES method [83]. More recently Yao et al. [84] proposed a new approach to characterize by LES the wind flow dynamics in the UCL, by parametrizing the atmospheric surface layer over a real dense city. The authors found that a rougher urban surface may generate eddies with a larger shear length scale, thus enhancing momentum transport.
While most of the publications dealing with UCL explore the in and above canopy regions, there is a range of scientific studies mainly focusing on the portion of wind flow developing only above the canopy (i.e. over idealized and real urban environments). Recently, Mo and Liu [85] conducted a study to characterize the wind flow in the ISL and RSL over a real urban district of Hong Kong city. Simulated data were then compared to wind-tunnel results for validation purpose. To better understand the turbulent structures of the wind flow developing over the urban city, simulations were carried out with the LES approach. In general, this is an important aspect commonly investigated in the research area of urban wind energy exploitation, as comprehensively reviewed by Toja-Silva et al. [86] and Jie et al. [87]. Back in 2016, Toja-Silva [88] proposed an empirical-heuristic optimization of the building-roof geometry for urban wind energy exploitation. However, these simulations were performed with RANS approach which is unable to properly predict regions of reversal and separation of the flow. For this reason, the same authors in the concluding section of that paper suggested the use of URANS and LES to better assess the unsteadiness of the flow. This is in line with the most recent advances on the subject, as confirmed by Kostadinović Vranešević et al. [89, 90] who used the LES approach to investigate the wind flow pattern around buildings with the aim to maximize energy potential resources.
Given the open-source nature of OpenFOAM, research studies have been also carried out with OpenFOAM to implement new numerical models (e.g. turbulence closure, inflow methods, solvers) and validate their performance on isolated and cluster of buildings in terms of wind flow velocity pattern and turbulence characteristics. RANS approach from OpenFOAM 1.6.x was used by Franke et al. [91] to simulate the wind flow around/over an array of simplified buildings, using different grid types (i.e. structured hexahedral, unstructured tetrahedral, unstructured hybrid grid of tetrahedra and prisms). The results were then validated by means of corresponding experimental results. Similarly, Wang et al. [92] carried out LES simulations to assess the wind flow pattern around an isolated simplified building with three grid types (i.e. hexahedral, tetrahedral and polyhedral cells). In general, they concluded that cases using polyhedral cells provided similar results to the case adopting hexahedral cells, and generally better than those cases using tetrahedral cells. A more extensive parametric study was conducted by Ozake et al. [93] on a 1:1:2 isolated building in the framework of the Architectural Institute of Japan [94]. Various computational grid arrangements, spatial discretization schemes, sub-grid scale (SGS) turbulence models and convergence criteria were tested with the LES. Also Liu et al. [95] and García-Sánchez et al. [96] investigated the performance of various turbulence modeling methods to predict airflow in simplified and realist urban environment. Ricci et al. [97] investigated the impact of turbulence models coupled to RANS approach and the impact of the surface roughness (in terms of equivalent sand-grain roughness height, ks) on the wind flow pattern in a realistic urban district. They found that the choice of the turbulence model may significantly influence the results in terms of wind speed and turbulent characteristics. The authors concluded that “… the choice of a specific turbulence model may be crucial for a large category of urban physics and wind engineering problems … although it is not a trivial issue to clearly and unambiguously define which one should be definitively used for a specific investigation …”. The setting of inflow CFD conditions is also crucial in research studies focusing on wind in built environments. Back in 2013, Juretić and Kozmar [98] proposed a novel method to prescribe inflow boundary conditions in RANS approach coupled with the standard k–ε turbulence model [99], to simulate a neutrally stratified ABL wind flow. The CFD results showed a good agreement with wind-tunnel data in terms of mean velocity, turbulent kinetic energy and Reynolds stress profiles. 3D steady RANS investigations aimed at understanding the impact of inflow ABL wind conditions were also conducted by Ricci et al. [100, 101]. The authors found that the choice of different neutral ABL wind conditions as well as different formulations from the literature [34, 47] to define the profiles of mean velocity and turbulence characteristics may lead to different results in the UCL. On this subject, García-Sanchez et al. [102] conducted a distinguished research study aimed at quantifying the uncertainties related to three inflow parameters: the wind velocity, the wind direction, the aerodynamic roughness length (z0) of the logarithmic profile. 3D steady RANS simulations were performed on a realistic urban environment. The uncertainty quantification method revealed a great potential to aid the development of more realistic inflow boundary conditions in atmospheric environment simulations. Unlike RANS simulations, as also stressed in Sect. 2 the definition of inflow conditions for LES simulations may be a challenging task, since it requires the choice of a specific inflow method to perturbate the flow field. Analyzing the numerous inflow methods developed in the past decade for LES simulations is beyond the scope of the present review paper. To support the reader, a non-exhaustive list of publications dealing with this topic is suggested here (e.g. [51, 103,104,105,106]). In these papers, the authors proposed novel inflow methods by comparing their performance with methods already available in the scientific literature for different type of subjects, from open terrains to isolated buildings. Among the others, Mansouri et al. [107] carried out an interesting study in OpenFOAM by testing four types of synthetic turbulence generator methods to provide a neutral ABL wind: digital filter methods, synthetic eddy methods with different shape functions, divergence free synthetic eddy method, and two types of anisotropy turbulent spot method. In general, the authors found that the four methods produced spurious pressure except for the cases using the synthetic eddy method with the Gaussian shape function.
Less traditional solvers (than RANS and LES) are also recently used to predict wind flows in the built environment. An example is offered by Xia et al. [108], who analyzed the performance of the Very-Large Eddy Simulation (VLES) approach [109, 110] to predict turbulent separated flows. In general, the results showed a good performance of the VLES when compared to available LES and experimental data. Li et al. [111] adopted the three fast fluid dynamics (FFD) models [112] with different pressure-correction schemes for solving Navier–Stokes equations in OpenFOAM: the standard incremental pressure-correction (SIPC), the non-incremental pressure-correction (NIPC), the rotational incremental pressure-correction (RIPC). The authors found that the SIPC is faster than NIPC and RIPC if using the same grids, turbulence model and time-step size. On this subject, more recently research studies have been conducted on built environments by using genetic algorithms and neural network methods for the optimization of urban wind conditions at local scales (e.g. [113,114,115,116]).
Overall, the OpenFOAM is largely used to investigate the wind flow development over/through built environments, but also to develop new models and solvers that can improve the prediction of wind characteristics around buildings, cluster of buildings and generally in/above the UCL. Assembling numerical benchmarks validated with experimental data is crucial to support the development of new analytical formulations of vertical wind profiles (velocity and turbulence) in complex environments.
3.2 Pedestrian wind comfort and urban microclimate
The pedestrian-level wind comfort as well as the thermal comfort are among the most investigated topics when dealing with the built environment. This is because the wind comfort and wind safety assessment of pedestrians is nowadays imperative in many countries when designing new buildings (mainly high-rise buildings) amidst historical and well-settled urban layouts. Strong winds in the middle of urban contexts can be extremely dangerous for people and medium/long-term impact on the economy of cities and communities. One of the first paper dealing with this topic was probably published by Jackson [117], who assessed the wind effects on pedestrians by using a pre-assembled database of measurements and questionary survey. Since then, various criteria to evaluate the wind comfort and safety were developed by academicians and practitioners, and numerous review papers on this topic were then published (e.g. [18, 118,119,120,121,122,123,124,125,126,127]). Pedestrian wind comfort studies can be grouped into thermal comfort studies and more simply wind comfort studies, defined by the author of this review as such when referring to the mechanical effect only (by neglecting the temperature variation) of the wind on pedestrians.
In general, the wind comfort studies represent about 52.2% of the comfort research studies (i.e. wind comfort and thermal comfort) here reviewed. About 67% of these wind comfort studies have been carried out with the 3D steady RANS approach. The reason behind this choice may be twofold. Firstly, as stressed by Blocken [8] best practice guidelines have been developed mainly for RANS simulations. Almost no guideline documents exist for LES and this is generally considered a serious limitation by CFD users. Secondly, as claimed in previous review papers (e.g. [30, 34]) the RANS approach can accurately predict high wind speed regions of flow which are mainly responsible for the discomfort. It means that, the inability of RANS in predicting low wind speed regions commonly does not lead to significant mispredictions in comfort analyses. On the other hand, the RANS approach usually requires sensitivity studies to understand how numerical parameters (e.g. geometrical simplification, grid size/type, boundary conditions, turbulence models, etc.) may affect the results. There are various examples of parametric studies on wind comfort carried out with OpenFOAM in the past years. Weerasuriya et al. [128] proposed a new inflow boundary condition to model twisted wind profiles generating from hilly topography upstream of urban environments. The research study was carried out with ANSYS Fluent and OpenFOAM. The authors observed that generally OpenFOAM was more reliable in predicting wind comfort with the newly proposed boundary conditions. Dhunny et al. [129] and Cindori et al. [130] conducted a sensitivity study to assess the pedestrian wind comfort in a tropical city and around an isolated (simplified) building, respectively, by investigating the impact of various k–ε and k–ω turbulence models coupled with RANS. Sensitivity studies with focus on different parameters were carried out by Tsichritzis and Nikolopoulou [131] and Hågbo et al. [132]. Tsichritzis and Nikolopoulou [131] analyzed the impact of morphological characteristics related to building height variation (in a real city) on the wind comfort. Hågbo et al. [132] investigated how different geometry inputs’ (to construct the computational grid) may influence the wind flow field at the pedestrian level. They found that the footprint extrusion model showed significant differences with respect to the national database and the Airborne LiDAR model, possibly affecting in turn the building design decisions in a well-settled urban layout (Fig. 5). However, performing CFD simulations on large-scale domains (like cities) to investigate the impact of several parameters can be very time consuming and sometimes prohibitive especially for practitioners. To overcome this issue, Kaseb et al. [133] proposed an innovative evolutionary-based framework to optimize building heights and plan area densities and improve wind comfort in complex real urban areas. The optimization method is based on a genetic algorithm and particle swarm optimization. The authors concluded that with this framework the computational efforts can be reduced significantly and the wind comfort can be improved in the urban area.
In general, the method adopted to define the mean wind speed in the wind comfort assessments may also play an important role, and this can change based on the CFD approach used for simulation. This was the goal of the LES study conducted by Kikumoto et al. [134], who focused on the mathematical interrelation among three mean wind speed types (i.e. mean-vector speed, mean speed, and effective speed) around an isolated building. The comparative analysis revealed that deviations may be significant near the ground surface and in recirculation regions of the flow. Ikegaya et al. [135] adopted the LES approach to simulate the wind flow around a 1:1:2 isolated building. Thus, they analyzed the probability density distribution for each velocity component and wind speed in order to provide a reliable model capable to predict the exceeding wind speeds at the pedestrian level. Likewise, Wang and Okaze [136] proposed a novel statistical method to estimate the low occurrence of strong wind speeds by the mean wind velocity and also the turbulent kinetic energy. In their study two case studies, previously simulated by Okaze et al. [137] with LES in OpenFOAM, were considered: an isolated building and a building array. Lin et al. [138] analyzed the wind comfort with the focus on the “downwash” and “Venturi effect”, previously observed experimentally by Tamura et al. [139]. For this purpose, authors carried LES simulations around square-section buildings with same aspect ratios and different sizes.
Mitigation methods aimed at reducing high wind speed zones in urban areas (i.e. to improve the comfort) are also generally conducted in CFD and validated by experimental data. As thoroughly reviewed by Buccolieri et al. [140], the vegetation (e.g. trees, grass, hedges) is one of the measures commonly used by engineers to improve the pedestrian-level wind comfort in the built environment. In OpenFOAM, the vegetation is often modeled as a porous medium and the solver porousSimpleFoam is therefore used for RANS simulations. However, a very limited number (i.e. three) of OpenFOAM studies [141,142,143] focusing on the mechanical effects (only) have been found by the author of this review paper, and two out of three are conference proceedings.
The thermal comfort studies represent about 47.8% of the total number of comfort studies reviewed in this paper. The majority of these deals with urban microclimate, urban heat island, natural ventilation and building energy simulation (BES). In the past decade, review papers have been published on these subjects (e.g. [144,145,146]) mainly focusing on urban microclimate and impact on the human health. In this regard, as for the wind comfort, mitigation measures have been developed both numerically and experimentally in order to improve the thermal comfort of people under adverse atmospheric conditions, especially in urban environments. The reader can find some interesting and exhaustive review papers recently published in the scientific literature (e.g. [147,148,149,150,151]). While OpenFOAM as well as ANSYS Fluent and ENVI-met [152] are the main CFD-based tools used to simulate atmospheric wind and temperature fields in urban environments, BES simulations for outdoor/indoor analyses are usually conducted with different tools (as EnergyPlus, ESP-r, SPARK). In microclimate analysis, the CFD results (in terms of wind, temperature, humidity, solar radiation, etc.) are generally used as input boundary conditions for BES simulations to determine the energy demand/performance. This process is commonly obtained by one-way CFD-BES coupling. As for the wind comfort, sensitivity studies on specific numerical/geometrical parameters are crucial also in urban microclimate. In general, research studies are primarily carried out on generic and simple geometries, often well-known benchmarks like canyons and building arrays. Once the impact of numerical parameters is well understood, more complex and realistic urban environments are then analyzed. Allegrini et al. [153] and Allegrini and Carmeliet [154] investigated the influence of different simplified urban morphologies (i.e. urban plan and building height) as well as of the buoyancy on the local microclimate. CFD simulations were performed with OpenFOAM while BES simulations were carried out with CitySim [155]. The results obtained in previous fundamental studies further motivated the authors to apply the one-way coupling OpenFOAM-CitySim on a real case scenario (a district of Zürich, Switzerland) to evaluate the increase of wind velocity and surface temperature due the presences of newly designed buildings amidst well-settled urban layout [156] (Fig. 6). Also Mokhatar and Reinhart [157] focused on a realistic scenario by investigating a district of San Francisco in California, United States. The authors proposed an innovative probabilistic, progressive, and accuracy-adaptive modeling approach for faster spatially resolved outdoor thermal comfort estimation. In this study, 3D steady RANS simulations were carried out with OpenFOAM by means of Eddy3D, a toolkit for decoupled outdoor thermal comfort simulations in urban areas [158]. Similary, a sensitivity study mainly focusing on the performance of different BES tools (starting from the same OpenFAOM CFD simulation) was carried out by Heidarinejad et al. [159] who investigated the influence of building surface solar irradiance on environmental temperatures in a simplified urban area. Urban microclimate studies may also be used to calibrate and monitor the performance of reference target buildings. This was the scope of the research conducted by King et al. [160] that analyzed the impact of several wind directions on a cubical building in isolation and in a staggered nine cube array format, with both OpenFOAM and ANSYS Fluent. The comparison was made using methods suggested by guide for ventilation based on internal/external building pressure differences. The authors observed that the ventilation rate of the cubical building in isolation (with single-sided opening) was lower than either of the cross-flow cases, and relationships between air change rate and wind angle were much weaker in the array cases. Miguel et al. [161] used OpenFOAM (for atmospheric simulation) and EnergyPlus (for building energy simulation) to simulate the outdoor conditions and the building energy performance of a reference building by evaluating potential countermeasures to urban heat island effects.
Overall, the two topics introduced in this subsection are based on a different workflow, different analyses, different tools and models. While for the wind comfort assessment the main factor is the wind speed and in few cases also the turbulence, in the thermal comfort coupling the wind and the temperature fields is essential. This latter can rely on a downscaling approach executed by means of open-source tools (e.g. WRF–OpenFOAM–EnergyPlus) with the potential to reach very small scales (e.g. the room scale) through BES simulations.
3.3 Pollution and gas dispersion
The World Health Organization estimates that pollutants and gases have caused about 4.2 million premature deaths around the world and about 99% of the world’s population breaths air exceeding the prescribed limits [162]. As stressed by the United Nations the percentage of the population living in urban area is rapidly increasing. Therefore, making cities more sustainable and resilient to climate change by lowering the carbon emission and increasing renewable energy resources is imperative [163, 164]. In the last two decades, great efforts to investigate the pollutant dispersion/concentration at different spatiotemporal scales have been made by the WE and Atmospheric Science community, under various atmospheric conditions and for a large variety of emitting sources (from traffic emissions to thermal power plants). As observed in Sect. 3.2, most of the research has been and is still conducted on the simplest and most common urban subsystem, that is the urban street canyon. The urban street canyon is an elementary configuration that through the years helped researchers to deeply understand the airflow and pollutant dynamics also in more complex urban environments (as cities) (e.g. [144, 165]). In general, the CFD technique has been widely used in this field to solve advection and diffusion equations of concentration based on the velocity field obtained from the Navier–Stokes equations. As opposite to wind tunnel and field measurements, the CFD allows the monitoring of output data (e.g. pressure, velocity, pollutant concentration, temperature, etc.) at any point of the domain, facilitating the understanding of complex flow dynamics throughout the entire volume/space under investigation (e.g. [8, 166]). Traffic emission sources like cars and trucks are usually modeled by means of ideal lines or points. Recently, an innovative research study has been carried out by Qin et al. [167] with ANSYS Fluent to understand how ideal sources can properly reproduce the emission traffic of realistic car source in terms of pollutant concentration in urban canyons. However, except for this study most of the RANS and LES studies are carried out on simplified urban street canyons and cities with ideal line/point sources, under neutral ABL wind conditions (i.e. with standard logarithmic wind velocity profile at the inlet) and creating thermal stratifications inside the domain (e.g. [168,169,170]). The thermal stratification is obtained by combining the desired reference wind velocity at the inlet and the surface heat flux on the surfaces (e.g. ground as well as buildings), through the evaluation of the Richardson dimensionless number. More complex scenarios than urban street canyon are also analyzed for different engineering purposes. Flores et al. [171] used DES and LES approaches to investigate the pollutant dispersion inside a simplified and realistic open pit mines under intense insolation. The authors found that buoyant currents may foster the exhaust of particles from the pit by modifying the mechanism of the flow recirculation inside the cavity and increasing the atmospheric turbulence. Reiminger et al. [172] used the RANS approach in combination with the Renormalization Group (RNG) k–ε turbulence model [173] to analyze the impact of wind speed and atmospheric stability (i.e. neutral, unstable and stable cases) on the pollution concentration within a street with noise barriers (i.e. like a street canyon).
On simplified urban street canyons, new models have been developed to execute CFD simulations on pollution and thermal stratification. Li et al. [174] proposed a new scheme based on simpleFoam (semi-implicit method for pressure linked equations) algorithm in OpenFOAM, the so-called TimeVarying-SIMPLE, to simulate the time-series flow field under time-varying inflows. The authors performed LES simulations on an ideal street canyon, and compared the wind flow and pollutant dispersion obtained with the new approach with those achieved with pisoFoam (pressure-implicit with splitting of operators) and pimpleFoam (the merge of PISO and SIMPLE) algorithms. Hang et al. [175] developed a new CFD code named APFoam to simulate the integrated processes of both complex photochemical pollutant reaction (NOx–O3–VOCs) and dynamic dispersion under two 2D street canyons with different aspect ratios, namely a typical street canyon and a deep street canyon (Fig. 7). Leng et al. [176] proposed an innovative micro-in-meso-scale simulation framework to downscale mesoscale meteorological data (obtained with WRF) at the microscale (by RANS simulations). Results are provided to the final users in terms of wind and pollutant concentration at the local scales (e.g. urban area or a district of it). Research studies using downscaling techniques (e.g. WRF—OpenFOAM) and focusing on gas leakage, gas dispersion, reactive pollutants and the related exposure risk for humans have also been found in the scientific literature (e.g. [177]). Most of these have been carried out on real industrial sites, as offshore platform for subsonic and sonic release [178], thermal power plants surrounded by urban area for comparison with analytical Gaussian plume models [179] as well as on simple and basic urban elements as street canyons [180].
As for wind comfort studies, vegetation (e.g. grass, trees, hedges) has been widely used to mitigate the pollutant concentration in urban and surrounding environments. However, as claimed by Janhäll [181] the “… design and choice of urban vegetation is crucial when using vegetation as an ecosystem service for air quality improvements. The reduced mixing in trafficked street canyons on adding large trees increases local air pollution levels, while low vegetation close to sources can improve air quality by increasing deposition…”. This is an important aspect investigated worldwide by CFD simulations, wind-tunnel testing and field measurements at different spatiotemporal scales. While the LES approach is mainly used for urban street canyons, the RANS approach combined to (various) k–ε and k–ω turbulence models is often preferred to simulate large-scale environments (as cities). It is worth noticing that, in this kind of studies the thermal stratification and buoyancy effects are rarely taken into account. Trees and hedges are the green elements more widely adopted and modeled in computational domains. Irrespective of the numerical approach used, trees and hedges in CWE can be (1) explicitly modeled with its components (usually trunk and branches) or (2) implicitly modeled as porous fluid zones. In the latter, an equivalent leaf area density and drag coefficient is assigned to the porous fluid zones and additional source/sink contributions are considered in transport equations [182]. Most of the studies uses the implicit modeling of vegetation. Vranckx et al. [183] investigated the impact of trees (implicitly modeled) on pollutant dispersion in street canyons by RANS and k–ε turbulence model, by using the porousSimpleFoam solver as the majority of RANS studies mentioned below. In line with the Janhäll [181], the authors observed an annual pollutant concentration increase of about 8% for element carbon and about 1.4% for PM10, due to the presence of vegetation. Different results were found by Jeanjean et al. [184,185,186], who investigated the potential of trees (implicitly modeled) as well as buildings facades coated with photocatalytic paint and solid barriers to disperse road traffic emissions in built environments. The cases of Leicester and London, in United Kingdom, were simulated by RANS and k–ε turbulence model. Overall, the authors found that trees may have a beneficial impact on road traffic emissions by reducing the ambient concentration of road traffic emissions of about 7% at pedestrian-level height.
More fundamental research studies were also conducted on this topic. Wang et al. [187] analyzed the performance of explicitly modeled trees in an urban street canyon by using the LES approach. In general, the authors found that pollutant concentration increases significantly with an accumulation of pollution between the trunk and the leeward side of the canyon. For larger inflow wind velocities, the pollutant concentration is generally heavier with respect to the level observed in the absence of trees. McMullan and Angelino [188] proposed a recycling and rescaling turbulent inflow generator method for LES, to assess the performance of trees on the traffic pollutant dispersion in a street canyon. The authors concluded that tree crowns may reduce the effectiveness of the canyon vortex in ventilating the street, by enhancing the mean tracer gas concentration at pedestrian level on the leeward wall. Regardless of numerical settings, the conclusions of Wang et al. [187] and McMullan and Angelino [188] are in line with the conclusions of Vranckx et al. [183], who noted an increase of pollutant concentration when introducing trees in the street canyon. This confirms that there is no general consensus on the efficiency of vegetation to reduce concentration in simple street canyons as well as in urban areas. Thus, further CFD studies are required to deeply understand the basic mechanisms of flow dynamics and pollutant deposition/dispersion under various thermal stratifications.
3.4 Wind-driven rain, windblown sand and snowdrift
This section focuses on the use of OpenFOAM to simulate the wind-driven rain, windblown sand and snowdrift in different environments. These three subtopics have been regrouped within the same subsection because similar numerical methods can be used to model particles.
The publications on wind-driven rain represent about 40.74% of the papers included in this subsection. The wind-driven rain is an important phenomenon largely investigated in the past decades by the WE community. This phenomenon may be responsible of soil and building materials erosion, degradation of building envelops as well as structural damages due to the moisture intrusion [189]. Back in 2004, Blocken and Carmeliet [190] conducted a comprehensive review by presenting the state-of-the-art of this topic in building science. The authors carefully reviewed the main experimental, semi-empirical and numerical methods adopted until then. The use of CFD was also reviewed with main focus on the RANS approach used in combination with k–ε turbulence models. In 2010 and 2022, Blocken and Carmeliet [191] and Gholamalipour et al. [192] presented a new review of the state-of-the-art. Regarding the numerical techniques, they discussed on the use of steady-state and transient CFD approaches (e.g. RANS and LES) as well as the applicability of Lagrangian particle tracking and Eulerian multiphase methods. In general, the Lagrangian and Eulerian methods are often combined with the RANS approach in wind-driven rain simulations, especially for built environments. The Lagrangian method can provide a clear and straightforward overview of the droplet-wall interaction. However, it requires a significant number of raindrops to achieve accurate results, by making this method computationally demanding. The Eulerian method treats the rain as a continuum (as for pollutant, gas, sand, snow), and for each rain class of raindrop size different rain phases are considered [189]. This method is generally way less computationally demanding than Lagrangian.
To the best knowledge of the author, one of the first research studies on wind-driven rain with OpenFOAM dates back 10 years and was conducted by Kulibaly et al. [193]. Here the authors investigated the potential negative impact of this phenomenon on the hygrothermal performance and durability of building facades of an array of low-rise buildings. For this study, the RANS approach was used with the RNG k–ε turbulence model. The Eulerian method was adopted for the spatiotemporal information of the wind-driven rain. CFD results were then validated by means of field measurements. The same case study was numerically reproduced by Kulibaly et al. [194] by LES and Eulerian method, and the simulate data were validated with reduced-scale wind-tunnel data. A similar methodology was then used by Kulibaly et al. [195] to simulate the impact of wind-driven rain on two parallel buildings (Fig. 8) that later became an experimental and numerical benchmark for this kind of studies (e.g. [196]). In particular, the same case study was reproduced also by Xu et al. [197] who proposed an innovative method to define an effective horizontal release plane for Lagrangian particle tracking of raindrops in space. The automatic selection of release plane is based on the analytical solution for unobstructed (empty domain) raindrop trajectories. A scaling factor is then adopted to account for the impact of any physical obstruction.
The windblown sand counts about 48.15% of the papers considered in this subsection. The windblown and sand accumulation can create dangerous and harmful conditions to the built environment at various levels, from industrial facilities to railways, making this topic of great interest for engineering disciplines, environmental sciences as well as in applied mathematics and physics [198]. As reported by Bruno et al. [198] who provided the WE community with a comprehensive review on this topic “… A growing demand for windblown sand mitigation design, building and maintenance has been observed in the last decade and it is expected to further increase in the next 20–30 years. The increasing interest in windblown sand mitigation is testified by the growing number of published studies and filed patents in the last years. …”. In this review the authors proposed an original categorization of windblown sand-induced performance deficiencies of the railway systems and prevention techniques to mitigate the windblown sand effects. Experimental (as wind tunnel) and numerical (as CFD) techniques are commonly used to conduct investigations in this field. In particular, the CFD technique played a fundamental role in the analysis of flow dynamics, sediment transport over/around structures (e.g. buildings) and infrastructures (e.g. sand barriers) as well as in aeolian geomorphology applications [199]. One of the first numerical studies on hills and embankments was performed under neutrally stratified ABL winds by Deaves [200, 201]. Zhang et al. [202] utilized the RANS approach in combination with the k–ε turbulence model to simulate the wind flow and investigate the wind-induced erosion of railway embankment. However, in both cases the authors did not provide any indication about the used CFD tools. To the best knowledge of the author, a limited amount of windblown sand studies in the WE literature were conducted with OpenFOAM. Bruno and Fransos [203] performed RANS simulations with shear–stress transport (SST) k–ω turbulence model [204] on sand transverse dune, in order to investigate the wind flow dynamics and the induced sand erosion, transport and deposition. Special focus was laid on the emerging three-dimensional coherent flow structures arising in the wake of the transverse dune. Bruno et al. [205] investigated the aerodynamic effects on the wind flow of innovative and more conventional sand mitigation solid barriers, by means of RANS and SST k–ω turbulence model. Design guidelines were provided through a new proposed empirical dimensionless performance estimator (Fig. 9). Mitigation measures were also systematically analyzed by Horvat et al. [206] who investigated the local wind flow and potential sedimentation patterns around the railways. The knowledge of sedimentation patterns is crucial for preventing sand-induced limit states that may produce in turn serious disruptions on the whole supply chain system. Wind flow field, shear stress field and derived sand sedimentation/erosion patterns were obtained by RANS in combination with the SST k–ω turbulence model. The performance of different railway substructures and track systems was discussed under different reference wind velocities and angles of attack. In general, the sediment transport has been investigated with OpenFOAM also for other types of environments. In particular, Pourteimouri et al. [207, 208] analyzed by RANS how the approach-wind direction and building spacing and shape may affect the wind flow and related sediment transport patterns at the beach. The authors observed that: (i) buildings’ location may have a positive effect on dune growth by steering the airflow and supplying sediment downstream; (ii) buildings can negatively affect the dune growth by obstructing the sediment from moving downstream. Hesp and Smyth [209] used the RANS and RNG k–ε model to simulate various dune scarps (with forward-facing steps), in order to understand the sediment transport pathways from beach to dunes in absence of structures. The authors demonstrated that the scarp morphology and the approach-wind direction can influence the transport pathways.
The snowdrift covers about 11.11% of total number of papers considered in this subsection. The snowdrift is the result of a very complex interaction between snow particle motion of three processes (i.e. creep, saltation and suspension) and wind blowing on/around the built environment. Investigating the spatiotemporal distribution of snow may help to prevent not only the disruption of services (with troubles to workers and inhabitants) but in the worst-case scenario also the collapse of the structures and infrastructures. Tominaga [210] conducted the first review in WE on CFD modeling of snowdrift around/on buildings later updated in 2018 [211]. As stressed by the authors, in snowdrift while the wind flow is modeled with the Eulerian method, two approaches can used for modeling the snow particle transport: the Lagrangian method or the Eulerian method. As mentioned above, the two methods have advantages and disadvantages. The Lagrangian method can predict the trajectory of individual particles but requires a large number of injected particles (computationally demanding) to obtain reliable statistics. The Eulerian method treats the snow particles as continuum transported passively by the airflow, through the mixture method (passive scalar transport equations) and the drift-flux method. Since 1990s various numerical models to simulate the three processes (i.e. creep, saltation and suspension) of the snowdrift on and around buildings have been proposed (e.g. [212,213,214,215]). Note that, these CFD studies were not carried out with OpenFOAM. In fact, a very limited number of research studies (only three) on snowdrift in built environment have been performed with OpenFOAM, while the majority found in the scientific literature was executed with ANSYS Fluent. One of the reasons may be that OpenFOAM is relatively “young” with respect to ANSYS Fluent. While Fluent in 1990s and 2000s was conventionally used for this kind of investigations, OpenFOAM was not born yet. To the best knowledge of the author, the first “published” research study on snowdrift executed with OpenFOAM dates to 2021 [216]. In this study, the authors performed RANS simulations (with the standard k–ε turbulence model) by adopting the Eulerian–Lagrangian method to quantify the small-scale snow deposition onto Arctic Sea ice. For this purpose, a new Eulerian–Lagrangian model so-called snowBedFoam 1.0 was implemented to simulate the snow transport. This model was developed based on the DPMFoam solver originally implemented in OpenFOAM 2.3.0. In the DPMFoam, the Eulerian continuum equations including particle volume fraction are solved for the fluid phase while Newton’s equations of motion are solved to calculate the trajectory of particles. As mentioned earlier, the coupling between wind and snowdrift can be a very complex phenomenon. In a short or even long period, the wind can cause the re-distribution of snow surface that will induce in turn a change on the approach wind field. To investigate this topic, Chen and Yu [217] proposed a dynamic mesh technology for the simulation of snowdrift to depict its surface evolution under long period. The driftUnsteadyScalarDyFoam was implemented in OpenFOAM and the results were validated by means of several measurement campaigns, including snowdrift on a flat roof and around a building. In order to improve the prediction of snowdrift in presence of obstacles, the same authors [218] developed a Lagrangian–Eulerian method, in particular the MPPICFoamEx solver (which embeds the pimpleFoam) was improved from the MPPICFoam solver. CFD simulations were aimed to assess the wind-induced snow transport on flat terrain and on a roadbed. A comparison with results from the Eulerian–Eulerian method (using the driftUnsteadyScalarDyFoam) and experimental tests was also provided in the paper. The authors concluded that the Eulerian–Lagrangian method provided a better agreement (than Eulerian-Eulerian approach) with the experimental data also in presence of obstacles (Fig. 10).
All CFD investigations related to windblown sand of this subsection were carried out with the steady RANS approach and not accounting for the particles deposition (e.g. [220]) since beyond the scope of the above-mentioned studies. On the contrary, for wind-driven rain as well as snowdrift studies unsteady simulations were also performed with Lagrangian–Eulerian and Eulerian–Eulerian methods. Nevertheless, as claimed by Horvat at el. [206] and remarked by Bruno, Benli and van Beeck in the mini-symposium Windblown Sand Sciences and Engineering of the 16th International Conference on Wind Engineering (ICWE16) [6], the windblown sand is an emerging issue of WE. There is still room for improvement, to use more sophisticated numerical approaches aimed at improving the prediction of sediment transport. The snowdrift topic has been largely investigated both for flat terrains and built environments (with buildings), with Eulerian–Eulerian and Eulerian–Lagrangian methods. However, this literature review revealed that just few “published” studies were carried out with OpenFOAM, while most of them were performed with ANSYS Fluent. For this purpose, new models/solvers were developed ad hoc and validated through experimental data by the users.
4 CFD for wind structural engineering
This section focuses on the use of OpenFOAM for wind structural engineering studies, which represent approximately the 29.79% of the total number of OpenFOAM studies of this review (Fig. 1a). In the past two decades, OpenFOAM has been widely used to investigate the aerodynamics and structural behavior of a large range of structures here grouped into five main sub-categories: bridges, low- and medium-rise buildings, high-rise buildings, lattice towers, photovoltaic (PV) solar panels. Note that, given the very large amount of research studies conducted with OpenFOAM on this topic, it was not possible to account for all of them. Thus, very comprehensive review papers are suggested here to support the reader. The section is organized into three subsections: wind effects on bridges (Sect. 4.1) with 9.04% of contributions, wind effects on low-rise and high-rise buildings (Sect. 4.2) with 17.02% of contributions, wind effects on PV panels (Sect. 4.3) with 3.72% of contributions (see Fig. 1b).
4.1 Wind effects on bridges
The cylinder (e.g. square and circular sections) is probably the simplest geometry, but also the most fascinating bluff body tested and simulated in the history of WE. This element has been used to understand the fundamental principles of the flow dynamics around bluff bodies, as the velocity and pressure field, the induced aerodynamic forces, the vortex-induced vibrations, and so on. It has been used and it is still used to mimic the behavior of more complex structures and/or its elements, as bridge decks, buildings, poles, etc. In 2008, the Benchmark on the Aerodynamics of a Rectangular 5:1 Cylinder (BARC) was announced during 6th International Colloquium on Bluff Body Aerodynamics and Applications (BBAA VI) hosted by Politecnico di Milano in Italy. Since then, scientists and scholars have been investigating the BARC for different purposes. More details about the research studies on BARC can be found in Bruno et al. [221] that carefully reviewed the extensive research activities and main results on this benchmark from 2010 to 2014. As mentioned above, the results achieved with the BARC may be representative of a wide range of bluff bodies, from bridge deck to sectional model of buildings, and this is probably the main reason this benchmark has been so widely used in the past years worldwide. Studies on cross-sectional bridge decks were carried out in OpenFOAM to investigate the static aerodynamic characteristics and flutter derivatives with URANS coupled to SST k–ω model (e.g. [222]) and LES approach (e.g. [223]). Andersen and Bossen [224] investigated a single bridge deck by URANS with SST k–ω model, with focus on the performance of three modal decomposition techniques for the identification of spatiotemporal structures (as vortex shedding effects) in the pressure field. Lateral barriers can also play an important role in the aerodynamics of bridge decks. Research studies on this subject are commonly conducted by wind-tunnel testing and CFD simulations to optimize the bridge and barrier performance. Investigating the performance of such elements can be extremely challenging by wind-tunnel tests, due to the scaling and the necessity to match similarity criteria. While, sensitivity studies can be performed on a large number of computational parameters (e.g. geometry, grid size/type, turbulence modeling, solvers, etc.) by CFD simulations in a reasonable time (e.g. [8, 27]). Li et al. [225] investigated by RANS (with SST k–ω model) the performance of a twin-winglet system used as an aerodynamic affiliated facility to increase the flutter stability of a box-girder suspension bridge. Xu et al. [226] used the 2D URANS approach to study the performance of porous elements positioned on bridge sections by following two strategies: (1) the pores are explicitly modeled; (2) barriers are implicitly modeled in terms of pressure-jump. The authors concluded that, although the strategy (2) has evident advantages compared to (1) since it allows saving computational resources and time, both strategies have strengths and limitations.
Numerical and experimental studies were also conducted with OpenFOAM on more complex sectional bridge models, as the twin-box bridge decks. Also in this case, parametric CFD studies on the gap distance between girders, on the transversal beams linking the deck girders, on the box geometry shape may be of great support at the early design stage of long-span bridges (e.g. [227,228,229]) (Fig. 11). Long- and medium-span bridges are also simulated in 3D, by reproducing the entire bridge. In particular, atmospheric wind conditions are downscaled from the mesoscale to the local scales (e.g. [230]) in order to understand its interaction with the surrounding environment (e.g. underlying terrain/waterway) as well as the aerodynamic performance of the bridge deck (e.g. [231, 232]). Less common but not less important are the research studies focusing on the effects caused by extreme winds on bridge decks. Hao and Wu [233] simulated the effect of a non-turbulent downburst wind field (modeled as an impinging jet) on a long-span bridge. CFD outputs were then used as dynamic input to the line-element-based 3D finite element model of the long-span bridge, and the downburst-induced transient response was acquired using the computational structural dynamics approach. Qin et al. [234] investigated the interactions between tornado-like vortices and a single-box deck with the LES approach, by focusing on the wind loads and main aerodynamic characteristics. With the LES approach, Chen et al. [235] analyzed the mechanism of the widespread failure of light poles on an elevated highway bridge exposed to Category 4 Hurricane Laura occurred in 2020.
Overall, it should be noted that the above-mentioned studies carried out under extreme wind conditions (as tornados, downbursts and hurricanes) do not represent an isolated case in the scientific literature, but on the contrary a new frontier. However, the numerical methods/models developed in the past decades (e.g. inflow methods for LES) to simulate the most common neutral ABL wind conditions are not always suitable to accurately reproduce highly transient winds as tornados and downbursts. Thus, further research studies are required on this subject to better understand the nature of these extreme winds, their impact on bridges, but also to improve the numerical technique. This aspect will be also discussed in Sects. 4.2 and 5.
4.2 Wind effects on low-rise and high rise structures
Buildings are the places where humans spend most of their time. Thus, the design of buildings has always been of primary importance for the WE community. The first benchmark examples of low-rise and high-rise buildings appeared in 1970s. The general scope of these benchmarks was to aid engineers in understanding the flow dynamics around buildings, wind-induced pressure and response of buildings under neutral ABL winds. For instance, the Commonwealth Advisory Aeronautical Research Council (CAARC) building [236] was probably the first benchmark of a single high-rise building and so far also the mostly investigated by the WE community, through experimental and numerical techniques. As stressed by Thordal et al. [24] who recently reviewed the numerical research activities on high-rise buildings, the CAARC has been widely used by scientists through the years to validate (with experimental data) models and settings for the wind loads assessment in structural design. In 1993, Mochida et al. [237] and Stathopoulos and Zhou [238] provided probably the very first numerical studies on subjects that would become shortly thereafter low-rise building benchmarks, for the analyses of wind flow field and wind-induced pressures by means of RANS and LES. The most remarkable aspect of this research activity is probably the use of LES approach “in 1993” for environmental/building simulations, by making the study of Mochida et al. [237] definitively a pioneering CWE research contribution. If the RANS approach in combination with k–ε and k–ω turbulence models has been for decades (from about 1970 to 2010) one of the most used CFD approach also in wind structural applications, the LES has become nowadays the most reliable and therefore widely adopted approach in this topic. This approach can provide users with more accurate turbulence characteristics (than RANS) and consequently a better solution in terms of pressure field, mean wind loads and peaks. However, running LES simulations on a cluster of buildings may be burdensome. In this regard, hybrid approaches have also been developed and used to simulate the wind flow field and loads on buildings. Back in 2011, the hybrid RANS—Kinetic Simulation approach was used by Huang et al. [239] to investigate the wind loads and pressure fluctuations on the CAARC. Similarly, Joubert et al. [240] and Wang and Chen [241, 242] adopted Spalart–Allmaras IDDES and the WMLES approaches, respectively, to investigate the complex time-varying three-dimensional flow field and wind loads on a high-rise building (referred as rectangular prism in [240]). Some studies related to low-rise and high-rise structures were also carried out with RANS and URANS. Nevertheless, these two approaches were either used (1) to simulate “only” the wind flow field around the bluff body for shape optimization process (e.g. [243,244,245]) without venturing a wind pressure/load assessment or (2) to develop machine learning/genetic algorithm frameworks for wind pressure/load assessment in combination with LES databases (e.g. [246,247,248]).
As mentioned earlier, the LES can provide a more accurate solution than RANS for wind load assessment, but the choice of the inflow method generator to reproduce neutral ABL wind conditions is crucial. A review on the numerous LES inflow method generators available in the literature as well as the validation studies and uncertainty quantification is beyond the scope of the present paper. The reader can find in the literature pivotal research studies carried out on open domains with different surface roughness classes, isolated low-rise buildings, high-rise buildings, slender structures and so on (e.g. [103,104,105, 249,250,251,252,253,254,255,256,257]). Note that, since these studies mainly focus on the inflow method generators OpenFOAM is not the only CFD tool adopted by the authors.
LES studies executed with OpenFOAM on low-rise and high-rise buildings in the structural engineering field can generally pursue multiple goals. Xing et al. [258] analyzed the accuracy and reliability of several computational parameters (i.e. grid size in proximity of the building, numerical schemes, inflow generator method) in LES simulation when predicting wind flow and wind loads on/around a real high-rise building, namely Torre Gioia 22, in Milan, Italy (Fig. 12). Marsland et al. [259] used the LES approach to investigate and reduce the wind-induced vibration by integrating the building with façade opening in service floors. Hasama et al. [260] evaluated the aerodynamic instability of a building through fluid–structure interaction analysis, by combining LES simulations and multi-degree-of-freedom structure model. Liu et al. [261] investigated with LES the performance of horizontal ribs, to be integrated on the facades of a high-rise building, with the aim to mitigate potential dangerous effects of strong winds. Melaku et al. [262] used the LES to assess the wind loads on a cube-shaped equipment mounted on the roof of a low-rise building. The variations of peak drag and uplift force coefficients with different equipment elevations were investigated in detail.
As for wind effects on bridges, in the past few years the increase of extreme meteorological events due to climate change led to a growing number of experimental and numerical studies focusing on the impact of windstorms (as downbursts, tornados, hurricanes) on low-rise structures mainly. This is because, while the ABL wind shows the highest wind speeds at the higher levels from the ground (more meaningful for tall structures), transient winds like downbursts and tornadoes manifest the highest wind speeds near the surface ground (e.g. [2]). Back in 2016, in one of the first CFD studies related to this topic and executed with OpenFOAM, Sim et al. [263] used the steady RANS coupled to the SST k–ω turbulence model to replicate a microburst-like wind as an impinging-jet flow. However few years later Žužul et al. [264], who carefully analyzed the reliability of URANS and Scale Adaptive Simulation (SAS) [265] to reproduce the complex dynamics of a reduced-scale downburst wind flows, recommended the use of more sophisticated approaches (like LES) to better resolve the complex turbulent structures of the flow (Fig. 13). Iida and Uematsu [266] used LES simulations to investigate the non-stationary mechanism causing strong winds close to the building walls and ground surface. With the same numerical approach, Cao et al. [267] simulated a tornado-like vortex to understand the wind-induced pressures and forces on a flat-roofed and gable-roofed low-rise building. The authors concluded that tornado-induced wind pressures have very different characteristics than those caused by synoptic winds like ABL flows.
A different type of structure that may also be classified as low-rise or high-rise structure (based on the height) are the lattice structures. For lattice structures, very few CFD studies conducted with OpenFOAM have been found in the literature. Nevertheless, given the peculiarity of such structures it may be worth to briefly discuss here some of these research studies. In general, these structures as well as many others have been investigated (e.g. [268]) under different wind conditions, and analytical models have also been derived for the wind-induced response assessment (e.g. [269,270,271,272]). One of the first CFD studies on lattice structures conducted in OpenFOAM probably dates back to 2014. It was carried out by Fabre et al. [273] on a real meteorological mast for which also wind-tunnel tests were performed at 1:20 reduced scale. The RANS simulations helped the authors to calculate the optimum anemometer position by avoiding any possible flow distortion. With respect to more conventional towers, these are peculiar structures generally composed by a large number of sub-elements. Generating a high-resolution computational grid around these structures requires advanced knowledge in geometry/grid modeling and also substantial computational resources, by making these studies commonly computationally demanding. Therefore, modeling only a portion of it to analyze local and most impacting wind effects is sometimes a reasonable “trade-off” for practical engineering applications. This was the strategy of Zhang et al. [274] and Hadane et al. [275] who adopted RANS and Wall-Adapting LES (WALE) approaches, respectively, to analyze the aerodynamic characteristics and wind loads on steel element of a transmission tower. The LES approach was used conversely by Ma and Sun [276] to simulate a hurricane boundary layer turbulence. The obtained wind flow field was then used to analyze the structural response of a power transmission tower.
4.3 Wind effects on photovoltaic panels
The use of renewable energy (as wind, solar and ocean) is crucial to achieve carbon neutrality and satisfy the target of net zero greenhouse gas emissions by the middle of twenty-first century (e.g. [277]). In this perspective, the use of photovoltaic (PV) panels increased significantly in the past years. PV panels are generally ground-mounted in open areas, rooftop-mounted and wall mounted on buildings. Besides their unquestionable and positive impact on the global ecosystem, PV panel systems may suffer from atmospheric and environmental factors, as the dust/particle deposition, rain drop as well as wind effects (e.g. [278]). As for the windblown sand, the investigation of wind effects on PV panels can be considered as an emerging topic in the WE community. As reported by Zhou et al. [278], who reviewed the CFD studies on the interaction between photovoltaic system and ambient environment, “…most studies applied CFD for simulations regarding PV thermal characteristics, PV cooling, and dust deposition and mitigation, whereas less were for investigating airflow and ventilation, wind loading, and microclimate…”. The present review paper confirms the observation made by Zhou et al. [278], since a limited amount of OpenFOAM studies focusing on this topic have been found in the scientific literature and only few of these deal with wind loading. Jubayer and Hangan [279, 280] adopted the RANS and URANS in combination with SST k–ω turbulence model to investigate the wind pressure coefficients, aerodynamic forces and flow structures on a stand-alone and an array of ground-mounted PV system, respectively. The results for the array configuration showed that the first windward row of PV panels experiences the maximum drag and lift forces. In terms of the maximum overturning moment, the 45° and 135° wind directions are the most critical with similar overturning moment coefficients for each row (Fig. 14). Similarly, Sun et al. [281] investigated ground-mounted PV panels by analyzing different scenarios in terms of wind velocities, arrangement of panel arrays like row and column spacing, panel inclination angle. A more sophisticated CFD approach was used by Wang et al. [282], who performed LES simulations for oblique wind directions (i.e. 45° and 135°) on flat-roof-mounted solar arrays to investigate the main dynamic characteristics of conical vortices and their effects on wind pressure distributions. Despite the increase of computational resources, performing sensitivity studies with LES to investigate the impact of numerical/environmental parameters can be burdensome and unpractical for such peculiar structures. To overcome this limitation, Aly and Clarke [283] proposed an innovative method combining RANS and machine learning methods to determine wind loads on PV panels. The study proved that this combination can significantly speed up simulations (up to 10,000 times faster) without losing in accuracy. As mentioned at the begin of this subsection, this is an emerging topic of the WE, where further numerical and experimental analysis are required to gain more knowledge on the wind flow dynamics and wind loads on/around such structures, in order to optimize the aerodynamic shape as well as the material selection/construction process.
Overall, the use of more sophisticated numerical approaches, like LES, is highly required in wind load assessment, if compared to topics analyzed in the previous sections. Although computationally demanding, this approach is essential when dealing with non-synoptic winds (as tornados and downbursts) in order to accurately solve the turbulent structures of the flow. The open-source nature of OpenFOAM can facilitate CFD users with the (1) development of new (LES) inflow method generators required to properly solve the complex turbulent structures and (2) the coupling of CFD simulations with FEM analysis.
5 Summary and perspectives
This review paper provided an overview on CFD research studies carried out by the WE community with the open-source OpenFOAM tool. The WE community and more specifically the CWE includes a wide range of topics and applications, spanning from pollutant/gas dispersion to wind structural engineering. A database of 289 papers, books and official websites has been assembled. Papers have been selected from well-known website platforms (as Science Direct) and by using the most common keywords appeared in the international WE and CWE conferences in the past decades. As expected, most of the papers (about 40%) were published on the reference journal of the WE and CWE community, namely Journal of Wind Engineering and Industrial Aerodynamics. Three main research areas have been identified in this review: (1) CFD for complex terrain: from mesoscale to microscale, (2) CFD for the built environment, (3) CFD for wind structural engineering. Within these research areas, various topics have been selected and discussed in the corresponding subsections.
The Wind Energy is one of the topics where OpenFOAM is most widely used. However, the Wind Energy can nowadays be considered a stand-alone subfield of Wind Engineering, with its own community and related international conferences worldwide. The same applies also for topics closely related to Wind Energy, dealing with wind and wave effects on off-shore structures (as off-shore/on-shore wind turbines, wave-structure interaction, floating structures, and so on). For this reason, only research studies closely related to wind energy exploitation in the built environment were considered in the present review.
The pre-processing of geometry and grid generation is always crucial to achieve successful CFD results, irrespective of the topic. For sake of brevity, although important in the present paper the author did not focus on geometry and grid generation procedures nor on the mostly used tools. Nevertheless, it may be interesting for the reader to know that the majority of the OpenFOAM studies discussed here does not always adopted grids generated with open-source tools/utilities of OpenFOAM (as SnappyHexMesh) or recently integrated in it (as CfMesh). On the contrary, in most of the cases either other open-source tools or commercial tools have been used to construct both geometry and grid. To some extent the lack of a powerful tool fully integrated into the OpenFOAM platform to generate and manipulate geometries and grids may represent a small limitation especially when dealing with wind flows in complex environments.
Similarly, in the past two decades there was a continuous development of the OpenFOAM code from different bodies, as OpenFOAM Foundation [284], OpenFOAM ESI [285], FOAM extend [286]. Analyzing the differences between these versions is out of the scope of the present review. Albeit these versions rely on the same basic algorithms, some differences can be found in terms of models/features/solvers availability by making some variants more versatile and suitable (than others) for specific WE applications. The choice of the version is therefore subjective and entrusted to the CFD user.
Since 2011, different CFD approaches have been used by wind engineers, from RANS to LES. Based on the topic, the CFD models have been often coupled to meteorological forecast models (as WRF), building energy simulation (BES) models, and finite element method (FEM) models. The possibility to couple OpenFOAM with other open-source platforms and establish operational framework procedures makes this tool undoubtably one of the most flexible and suitable CFD tools for research and practical engineering applications.
RANS has been the mostly used numerical approach in the early stage of OpenFOAM, but the rapid increase of computational resources of the last years made the LES one of the mostly used approach in WE nowadays. Nevertheless, RANS remains still the preferred approach for mesoscale simulations (on large-scale domains), pedestrian-level wind comfort assessment (when focusing on areas with high amplification factors, which are often the areas of primary interest in those studies), wind-driven rain, windblown sand and snowdrift (with Eulerian-Eulerian or Eulerian-Langrangian method). The large amount of sensitivity research studies (i.e. on inflow conditions, space and time discretization, turbulence models, algorithm solvers, etc.) carried out through the years with RANS have been crucial to improve the reliability of this approach and to favor in turn the development of best practice guidelines for CFD simulations in WE. On the contrary, there are WE topics for which RANS provides inaccurate solutions (due to well-known limitations). This aspect is particularly sensitive in research studies dealing with wind load assessment on structures (as bridges and buildings), for which the knowledge of the complex turbulent structures of the flow and peak values (velocity and pressure) is crucial both for neutral ABL winds and extreme winds (as downbursts and tornados). Hybrid RANS-LES approaches (e.g. DES, DDES, IDDES, WLES, SAS, etc.) may also represent an alternative option in case of large-scale domains (i.e. large number of control volumes) if the turbulent structures of the flow are sufficiently resolved.
Various inflow generator methods have been developed through the years for LES in OpenFOAM (and other CFD tools) to simulate neutral ABL wind flows with/without buildings. These are essential to properly develop the wind profiles and turbulence characteristics within the domain. However, although these are generally well-established for neutral ABL winds, their reliability to properly reproduce extreme winds (as downbursts and tornados) is still an open question for which more research is needed. To the best knowledge of the author this has a twofold explanation. (1) Tornados and downbursts are more complex to model (because of their transient nature) and validate by experimental data (barely available in the literature). (2) The occurrence of extreme events massively increased in the last years due to the climate change and the WE community is quickly adapting to this change, by focusing more on extreme winds rather than well-known neutral ABL winds. However, while WE standards and codes are fully based on a rational approach (i.e. so-called “Davenport chain”) that ties together synoptic winds (as ABL wind) and wind effects on structures, a similar rational scheme that joins the non-synoptic winds (as thunderstorms) and wind loading into a homogeneous framework is still missing. As stated by Solari et al. [287] “… this happens because the complexity of thunderstorms makes it difficult to establish physically realistic and simple models. Their short duration and small size make few data available. There is a persistent gap between wind engineering and atmospheric sciences …”. In the past few years, the scientific community made a big step forward in the measuring, simulating and modeling of extreme winds for WE applications. A unique initiative to enable a better understanding of fundamental downburst flow features relevant for the structural loading was framed under the umbrella of the European Research Council (ERC) Advanced project THUNDERR [288]. The scope of this project was absolutely novel: “… to detect thunderstorms, create a database of meteorological records and scenarios, conduct unprecedented laboratory tests and numerical simulations, formulate a thunderstorm model appropriate for both atmospheric sciences and structural design, change the format of wind actions, engineering practice and the codification system, make building safer and more sustainable, bring about a profound impact on society and its economy. …”. The research outputs of this project made clear that the open-source nature of weather models (as WRF), CFD codes (as OpenFOAM) and measurements is essential to understand the genesis of such winds. From the computational perspective, a large number of LES simulations under different scenarios are usually required to properly calibrate the model and understand the physics. Therefore, parallelizing the simulations on multiple nodes through HPC systems without paying any license cost may be very beneficial for the CFD user. Another important aspect concerns the coupling of weather forecasting models (necessary for mesoscale simulations) with OpenFOAM CFD simulation of local scales (e.g. buildings). This process may be easier and less costly because of the open-source nature of the utilized codes. Thus, if on the one hand OpenFOAM requires a relatively long training period and specific programming skills (than other commercial CFD tools), the long-term benefit for the CFD user and stakeholders is priceless.
Advanced technologies as artificial intelligence (AI) based schemes (e.g. machine learning) can aid the WE community with the better understanding of physical phenomena [26]. While such advanced technologies are widely adopted by the Fluid Dynamics/Mechanics as well as the Atmospheric Science communities, a limited number of WE studies (about 5.8% of those analyzed here) combining OpenFOAM and AI-based schemes have been found in the scientific literature. Nevertheless, this rate does not look that pessimistic, since about 90.9% of these studies have been recently published (from 2019 to 2023) by indicating an increasing confidence of CWE users in coupling AI-based schemes and OpenFOAM. This is also confirmed by Lamberti and Gorlé [236] on the use of a multi-fidelity machine learning framework to predict wind loads on buildings “… these results demonstrate that multi-fidelity machine learning frameworks provide a route to balancing accuracy and computational cost in the prediction of complex turbulent flow quantities”. In general, if on the one hand the promising results confirm that the CWE community is ready to embrace the AI revolution, on the other hand a “conscious use” of these new technologies in combination with experimental and numerical methods is imperative.
In conclusion, OpenFOAM plays and will play in the coming years a major role in the CWE community. In a totally unique way, it allows CFD users to access and manipulate source codes as well as implement new models/solvers with the support of the theory guide and the OpenFOAM community. OpenFOAM represents an extraordinary opportunity for all CFD users worldwide to share codes and exchange case studies (also through the OpenFOAM journal [289]), to create benchmarks for the development of new best practice guidelines, to explore the potential of new functionalities, to strengthen the network within the CFD community.
Data availability
Not applicable.
References
Cermak JE (1975) Applications of fluid mechanics to wind engineering—a Freeman scholar lecture. J Fluids Eng 97:9–38
Solari G (2019) Wind science and engineering. Origins, develpoments, fundamentals and advancements, 1st edn. Springer. ISBN-10: 3030188175
Solari G (2020) Education and dissemination in wind science and engineering. J Wind Eng Ind Aerodyn 203:104241
Solari G (1998) Wind engineering: a short introduction. Meccanica 33:213–322
Owen J, Sterling M, Hargreaves D, Baker CJ (2013) Fifty years of wind engineering: prestige lectures from the sixth European and African conference on wind engineering. University of Birmingham, UK
ICWE 16. The 16th international conference on wind engineering. https://www.icwe2023.com/topics/. Accessed 02 May 2024
Baker CJ (2007) Wind engineering—past, present and future. J Wind Eng Ind Aerodyn 95:843–870
Blocken B (2015) Computational fluid dynamics for urban physics: importance, scales, possibilities, limitations and ten tips and tricks towards accurate and reliable simulations. Build Environ 91:219–245
Hangan H, Refan M, Jubayer C, Romanic D, Parvu D, LoTufo J, Costache A (2017) Novel techniques in wind engineering. J Wind Eng Ind Aerodyn 171:12–33
Blocken B (2014) 50 years of computational wind engineering: past, present and future. J Wind Eng Ind Aerodyn 129:69–102
Davenport AG (1991) Preface to the proceedings of the 8th international conference on wind engineering, London, Ontario
Murakami S (1990) Preface to special issue on computational wind engineering. J Wind Eng Ind Aerodyn 35(1):ix–xi
Murakami S (1997) Current status and future trends in computational wind engineering. J Wind Eng Ind Aerodyn 67–68:3–34
Meroney RN (1997) Preface to special issue of second international symposium on computational wind engineering. J Wind Eng Ind Aerodyn 67–68:vii–viii
Stathopoulos T (1997) Computational wind engineering: past achievements and future challenges. J Wind Eng Ind Aerodyn 67–68:509–532
Tamura T (1999) Reliability on CFD estimation for wind-structure interaction problems. J Wind Eng Ind Aerodyn 81(1–3):117–143
Tamura T (2008) Towards practical use of LES in wind engineering. J Wind Eng Ind Aerodyn 96(10–11):1451–1471
Mochida A, Lun IYF (2008) Prediction of wind environment and thermal comfort at pedestrian level in urban area. J Wind Eng Ind Aerodyn 96(10–11):1498–1527
Murakami S, Matsumoto M, Tamura Y (2008) Preface to the special issue of the fifth symposium on computational wind engineering. J Wind Eng Ind Aerodyn 96(10–11):1449–1450
Fernando HJS (2010) Fluid dynamics of urban atmospheres in complex terrain. Annu Rev Fluid Mech 42:365–389
Barlow JF (2014) Progress in observing and modelling the urban boundary layer. Urban Clim 10:216–240
Löhner R, Haug E, Michalski A, Muhammad B, Drego A, Nanjundaiah R, Zarfam R (2015) Recent advances in computational wind engineering and fluid–structure interaction. J Wind Eng Ind Aerodyn 144:14–23
Tamura Y, Van Phuc P (2015) Development of CFD and applications: Monologue by a non-CFD-expert. J Wind Eng Ind Aerodyn 144:3–13
Thordal MS, Bennetsen JC, Koss HHH (2019) Review for practical application of CFD for the determination of wind load on high-riose buildings. J Wind Eng Ind Aerodyn 186:155–168
Kataoka H, Ono Y, Enoki K (2020) Applicatios and prospects of CFD for wind engineering fields. J Wind Eng Ind Aerodyn 205:104310
Kareem A (2020) Emerging frontiers in wind engineering: computing, stochastics machine learning and beyond. J Wind Eng Ind Aerodyn 206:104320
Potsis T, Tominaga Y, Stathopoulos T (2023) Computational wind engineering: 30 years of research progress in building structures and environment. J Wind Eng Ind Aerodyn 234:105346
Stathopoulos T (2002) The numerical wind tunnel for industrial aerodynamics: Real or virtual in the new millennium? Wind Struct 5(2–4):193–208
Hanjalic K (2005) Will RANS survive LES? A view of perspectives. J Fluid Eng 127:831–839
Blocken B (2018) LES over RANS in building simulation for outdoor and indoor applications: A foregone conclusion? Build Simul 11:821–870
Casey M, Wintergerste T (2000) Best practice guidelines. ERCOFTAC Special Interest Group on Quality and Trust in Industrial CFD, ERCOFTAC, Brussels
Franke J, Hellsten A, Schlünzen H, Carissimo B (2007) Best practice guideline for the CFD simulation of flows in the urban environment. COST Office Brussels, ISBN 3–00–018312–4
Britter R, Schatzmann M (2007) Model evaluation guidance and protocol document COST action 732. COST Office Brussels, 3–00–018312–4
Tominaga Y, Mochida A, Yoshie R, Kataoka H, Nozu T, Yoshikawa M, Shirasawa T (2008) AIJ guidelines for practical applications of CFD to pedestrian wind environment around buildings. J Wind Eng Ind Aerodyn 96(10–11):1749–1761
Blocken B, Gualtieri C (2012) Ten iterative steps for model development and evaluation applied to computational fluid dynamics for environmental fluid mechanics. Environ Model Softw 33:1–22
Roache PJ (1994) Perpective—a method for uniform reporting of grid refinement studies. J Fluids Eng Trans 116(3):405–413
Roache PJ (1997) Quantification of uncertainty in computational fluid dynamics. Annu Rev Fluid Mech 29:123–160
AIAA (1998) Guide for the verification and validation of computational fluid dynamics simulations. American Institute of Aeronautics and Astronautics, AIAA, Reston, VA (AIAA-G-077–1998)
Oberkampf WL, Trucano TG, Hirsch C (2004) Verification, validation, and predictive capability in computational engineering and physics. Appl Mech Rev 57(5):345–384
Tominaga Y, Liangzhu W, Zhiqiang Z, Stathopoulos T (2023) Accuracy of CFD simulations in urban aerodynamics and microclimate: Progress and challenges. J Wind Eng Ind Aerodyn 243:110723
ANSYS Inc. (2018) ANSYS fluent—theory guide 19.2
Weller HG, Tabor G, Jasak H, Fureby C (1998) A tensorial approach to computational continuum mechanics using object-oriented techniques. Comput Phys 12:620–631
Jasak H (2009) OpenFOAM: open source CFD in research and industry. Int J Nav Archit Ocean Eng 1(2):89–94
Orlanski I (1975) A rationale subdivision of scales for atmospheric processes. Bull Am Meteorol Soc 56(5):527–530
Ren H, Laima S, Chen W-L, Zhang B, Guo A, Li H (2018) Numerical simulation and prediction of spatial wind field under complex terrain. J Wind Eng Ind Aerodyn 180:49–65
Cheynet E, Liu S, Ong MC, Jakobsen JB, Snaebjornsson J, Gatin I (2020) The influence of terrain on the mean wind flow characteristics in a fjord. J Wind Eng Ind Aerodyn 205:104331
Richards P, Hoxey R (1993) Appropriate boundary conditions for computational wind engineering models using the k–ε turbulence model. J Wind Eng Ind Aerodyn 46:145–153
Balogh M, Parente A, Benocci C (2012) RANS simulation of ABL flow over complex terrains applying an enhanced k–ε model and wall function formulation: implementation and comparison for fluent and OpenFOAM. J Wind Eng Ind Aerodyn 104–106:360–368
Ricci A, Burlando M, Repetto MP, Blocken B (2022) Static downscaling of mesoscale wind conditions into an urban canopy layer by a CFD microscale model. Build Environ 225:109626
Deodatis G (1996) Simulation of ergodic multivariate stochastic processes. J Eng Mech 122(8):778–787
Melaku AF, Bitsuamlak GT (2021) A divergence-free inflow turbulence generator using spectral representation method for large-eddy simulation of ABL flows. J Wind Eng Ind Aerodyn 212:104580
Gopalan H (2018) Evaluation of Wray–Agarwal turbulence model for simulation of neutral and non-neutral atmospheric boundary layers. J Wind Eng Ind Aerodyn 182:322–329
Toparlar Y, Blocken B, Maiheu B, van Heijst G (2019) CFD simulation of near-neutral atmospheric boundary layer: new temperature inlet profile consistent with wall functions. J Wind Eng Ind Aerodyn 191:91–102
Xie ZT, Castro IP (2008) Efficient generation of inflow conditions for large eddy simulation of street-scale flows. Flow Turbul Combust 81(3):449–470
Vonlanthen M, Allegrini J, Carmeliet J (2017) Multiscale interaction between a cluster of buildings and the ABL developing over a real terrain. Urban Clim 20:1–19
Miao Y, Liu S, Chen B, Zhang B, Wang S, Li S (2013) Simulating urban flow and dispersion in Beijing by coupling a CFD model with the WRF model. Adv Atmos Sci 30(6):1663–1678
Skamarock WC, Klemp JB, Dudhia J, Gill DO, Liu Z, Berner J, Wang W, Powers JG, Duda MG, Barker DM, Huang X-Y (2021) A description of the advanced research WRF model version 4.3. NCAR/TN-556+STR. https://doi.org/10.5065/1dfh-6p97
Wong NH, He Y, Nguyen NS, Raghavan S, Martin M, Hii DJC, Yu Z, Deng J (2021) An integrated multiscale urban microclimate model for the urban thermal environment. Urban Clim 35:100730
Kadaverugu R, Purohit V, Matli C, Biniwale R (2021) Improving accuracy in simulation of urban wind flows by dynamic downscaling WRF with OpenFOAM. Urban Climate 38:109912
Castorrini A, Gentile S, Geraldi E, Bonfiglioli A (2021) Increasing spatial resolution of wind resource prediction using NWP and RANS simulation. J Wind Eng Ind Aerodyn 210:104499
Che Y, Salazar AA, Peng S, Zheng J, Chen Y, Yuan L (2022) A multi-scale model for day-ahead wind speed forecasting: a case study of the Houhoku wind farm, Japan. Sustain Energy Technol Assess 52:101995
Temel O, van Beeck JPAJ (2016) Adaptation of mesoscale turbulence parameterization schemes as RANS closures for ABL simulations. J Turbul 17(10):966–997
Taylor PA, Teunissen HW (1985) The Askervein Hill project: report on the September/October 1983 main field experiment. Technical Report MSRB-84–86, Atmospheric Environment Service, Downsview, ON, Canada
Temel O, Bricteux L, van Beeck JPAJ (2018) Coupled WRF-OpenFOAM study of wind flow over complex terrain. J Wind Eng Ind Aerodyn 174:152–169
Temel O, Porchetta S, Bricteux L, van Beeck JPAJ (2018) RANS closures for non-neutral microscale CFD simulations sustained with inflow conditions acquired from mesoscale simulations. Appl Math Model 53:635–652
Allwine KJ, Flaherty JE (2006) Joint urban 2003: study overview and instrument locations. PR 15967, PNNL
García-Sánchez C, Gorlé C (2018) Uncertainty quantification for microscale CFD simulations based on input from mesoscale codes. J Wind Eng Ind Aerodyn 176:87–97
Mughal MO, Lynch M, Yu F, Sutton J (2018) Forecasting and verification of winds in an East African complex terrain using coupled mesoscale—and micro-scale models. J Wind Eng Ind Aerodyn 176:13–20
Piroozmand P, Mussetti G, Allegrini J, Mohammadi MH, Akrami E, Carmeliet J (2020) Coupled CFD framework with mesoscale urban climate model: application to microscale urban flows with weak synoptic forcing. J Wind Eng Ind Aerodyn 197:104059
Castorrini A, Gentile S, Geraldi E, Bonfiglioli A (2023) Investigations on offshore wind turbine inflow modelling using numerical weather prediction coupled with local-scale computational fluid dynamics. Renew Sustain Energy Rev 171:113008
Kc A, Whale J, Urmee T (2019) Urban wind conditions and small wind turbines in the built environment: a review. Renew Energy 131:268–283
Ricci A, Burlando M, Freda A, Repetto MP (2017) Wind tunnel measurements of the urban boundary layer development over a historical district in Italy. Build Environ 111:192–206
Britter RE, Hanna SR (2003) Flow and dispersion in urban areas. Annu Rev Fluid Mech 35:469–496
Meroney RN (2016) Ten questions concerning hybrid computational/physical model simulation of wind flow in the built environment. Build Environ 96:12–21
Zhao Y, Chew LW, Fan Y, Gromke C, Hang J, Yu Y, Ricci A, Zhang Y, Xue Y, Fellini S, Mirzaei PA, Gao N, Carpentieri M, Salizzoni P, Niu J, Carmeliet J (2023) Fluid tunnel research for challenges of urban climate. Urban Clim 51:101659
Flores F, Garreaud R, Muñoz RC (2013) CFD simulations of turbulent buoyant atmospheric flows over complex geometry: solver development in OpenFOAM. Comput Fluids 82:1–13
Carpentieri M, Robins AG (2015) Influence of urban morphology on air flow over building arrays. J Wind Eng Ind Aerodyn 145:61–74
Ricci A, Kalkman I, Blocken B, Burlando M, Freda A, Repetto MP (2017) Local-scale forcing effects on wind flows in an urban environment: impact of geometrical simplifications. J Wind Eng Ind Aerodyn 170:238–255
Paden I, García-Sanchez C, Ledoux H (2022) Towards automatic reconstruction of a 3D city models tailored for urban flow simulations. Front Built Environ. https://doi.org/10.3389/fbuil.2022.899332
Furtak-Cole E, Ngan K (2020) Predicting mean velocity profiles inside urban canyons. J Wind Eng Ind Aerodyn 207:104280
Spalart P, Deck S, Shur M, Squires K, Strelets M, Travin A (2006) A new version of detached-eddy simulation, resistant to ambiguous grid densities. Theor Comput Fluid Dyn 20:181–195
Shur M, Spalart P, Strelets M, Travin A (2008) A hybrid RANS-LES approach with delayed DES and wall-modelled LES capabilities. Int J Heat Fluid Flow 29(6):1638–1649
Mockett C (2009) A comprehensive study of detached-eddy simulation. Ph.D thesis. Technische Universität Berlin
Yao L, Liu CH, Brasseur GP, Chao CYH (2023) Wind and eddy dynamics in the urban canopy layer over a city: a parametrization based on the mixing-layer analogy. Build Environ 246:110962
Mo Z, Liu CH (2023) Inertial and roughness sublayer flows over real urban morphology: a comparison of wind tunnel experiment and large-eddy simulation. Urban Clim 49:101530
Toja-Silva F, Kono T, Peralta C, Lopez-Garcia O, Chen J (2018) A review of computational fluid dynamics (CFD) simulations of the wind flow around buildings for urban wind energy exploitation. J Wind Eng Ind Aerodyn 180:66–87
Jie P, Su M, Gao N, Ye Y, Kuang X, Chen J, Li P, Grunewald J, Xie X, Shi X (2023) Impact of urban wind environment on urban building energy: a review of mechanisms and modeling. Build Environ 245:110947
Toja-Silva F, Lopez-Garcia O, Peralta C, Navarro J, Cruz I (2016) An empirical–heuristic optimization of the building-roof geometry for urban wind energy exploitation on high-rise buildings. Appl Energy 164:769–794
Kostadinović Vranešević KK, Vita G, Bordas SPA, Šarkić Glumac A (2022) Furthering knowledge on the flow pattern around high-rise buildings: LES investigation of the wind energy potential. J Wind Eng Ind Aerodyn 226:105029
Kostadinović Vranešević K, Ćorić S, Šarkić Glumac A (2023) LES study on the urban wind energy resources above the roof of buildings in generic cluster arrangements: impact of building position. J Wind Eng Ind Aerodyn 240:105503
Franke J, Sturm M, Kalmbach C (2012) Validation of OpenFOAM 1.6.x. with the German VDI guideline for obstacle resolving micro-scale models. J Wind Eng Ind Aerodyn 104–106:350–359
Wang W, Cao Y, Ozake T (2021) Comparison of hexahedral, tetrahedral and polyhedral cells for reproducing the wind field around an isolted building by LES. Build Environ 195:107717
Ozake T, Kikumoto H, Ono H, Imano M, Ikegaya N, Hasama T, Nakao K, Kishida T, Tabata Y, Nakajima K, Yoshie R, Tominaga Y (2021) Large-Eddy simulation of flow around an isolated building: a step-by-step analysis of influencing factors on turbulent statistics. Build Environ 202:108021
Architectural Institue of Japan (AIJ). https://www.aij.or.jp/eng/database/index_db.html. Accessed 02 May 2024
Liu J, Heidarinejad M, Pitchurov G, Zhang L, Srebric J (2018) An extensive comparison of modified zero-equation, standard k–ε, and LES models in predicting urban airflow. Sustain Cities Soc 40:28–43
García-Sanchez C, van Beeck van Beeck JPAJ, Gorlé C (2018) Predictive large eddy simulations for urban flows: challenges and opportunities. Build Environ 139:146–156
Ricci A, Kalkman I, Blocken B, Burlando M, Repetto MP (2020) Impact of turbulence models and roughness height in 3D steady RANS simulations of wind flow in an urban environment. Build Environ 171:106617
Juretić F, Kozmar H (2013) Computational modeling of the neutrally stratified atmospheric boundary layer flow using the standard k–ε turbulence model. J Wind Eng Ind Aerodyn 115:112–120
Jones W, Launder B (1972) The prediction of laminarization with a two-equation model of turbulence. Int J Heat Mass Transf 15(1972):301–314
Ricci A, Kalkman I, Blocken B, Burlando M, Freda A, Repetto MP (2018) Large-scale forcing effects on wind flows in the urban canopy: impact of inflow conditions. Sustain Cities Soc 42:593–610
Ricci A, Burlando M, Repetto MP, Blcoken B (2019) Simulation of urban boundary layer and canopy layer flows in port areas induced by different marine boundary layer inflow conditions. Sci Total Environ 670:876–892
García-Sanchez C, Philips DA, Gorlé C (2014) Quantifying inflow uncertainties for CFD simulations of the flow in downtown Oklahoma City. Build Environ 78:118–129
Patruno L, Ricci M (2018) A systematic approach to the generation of synthetic turbulence using spectral methods. Comput Method Appl Mech Eng 340:881–904
Vasaturo R, Kalkman I, Blocken B, van Wesemael PJV (2018) Large eddy simulation of the neutral atmospheric boundary layer: performance evaluation of three inflow methods for terrains with different roughness. J Wind Eng Ind Aerodyn 173:241–261
Potsis T, Stathopoulos T (2022) A novel computational approach for an improved expression of the spectral content in the lower atmospheric boundary layer. Buildings 12:788
Abu-Zidan Y, Nguyen K (2023) A machine learning approach for calibrating ABL profiles in large-eddy simulations. J Wind Eng Ind Aerodyn 232:105277
Mansouri Z, Selvam RP, Chowdhury AG (2022) Performance of different inflow turbulence methods for wind engineering applications. J Wind Eng Ind Aerodyn 229:105141
Xia Z, Cheng Z, Han X, Mao J (2020) VLES turbulence modelling for separated flow simulation with OpenFOAM. J Wind Eng Ind Aerodyn 198:104077
Han X, Krajnovic S (2013) An efficient very large eddy simulation model for simulation of turbulent flow. Int J Numer Methods Fluids 71:1341–1360
Han X, Krajnovic S (2015) Very-large eddy simulation based on k–ω model. AIAA J 53:1103–1108
Li R, Liu Z, Feng Z, Gao N (2022) Fast fluid dynamics simulation of the airflow distributions in urban residential areas. Energy Build 255:111635
Stam J (1999) Stable fluids. In: Proceedings of the 26th annual conference on computer graphics and interactive techniques, pp 121–128
Kaseb Z, Rahbar M (2022) Towards CFD-based optimization of urban wind conditions: comparison of genetic algorithm, particle swarm optimization, and a hybrid algorithm. Sustain Cities Soc 77:103565
Kastner P, Dogan T (2023) A GAN-based surrogate model for instantaneous urban wind flow prediction. Build Environ 242:110384
Liu Z, Zhang S, Shao X, Wu Z (2023) Accurate and efficient urban wind prediction at city-scale with memory-scalable graph neural network. Sustain Cities Soc 99:104935
Shao X, Liu Z, Zhang S, Zhao Z, Hu C (2023) PIGNN-CFD: a physics-informed graph neural network for rapid predicting urban wind field defined on unstructured mesh. Build Environ 232:110056
Jackson PS (1978) The evaluation of windy environments. Build Environ 13:251–260
Ratcliff MA, Peterka JA (1990) Comparison of pedestrian wind acceptability criteria. J Wind Eng Ind Aerodyn 36:791–800
Bottema M (2000) A method for optimisation of wind discomfort criteria. Build Environ 35:1–18
Blocken B, Carmeliet J (2004) Pedestrian wind environment around buildings: literature review and practical examples. J Therm Envel Build Sci 28:107–159
Stathopoulos T, Wu H, Zacharias J (2004) Outdoor human comfort in an urban climate. Build Environ 39:297–305
Sanz-Andres A, Cuerva A (2006) Pedestrian wind comfort: feasibility study of criteria homogenization. J Wind Eng Ind Aerodyn 94:799–813
Blocken B, Stathopoulos T (2013) CFD simulation of pedestrian-level wind conditions around buildings: past achievements and prospects. J Wind Eng Ind Aerodyn 121:138–145
Blocken B, Stathopoulos T, van Beeck JPAJ (2016) Pedestrian-level wind conditions around buildings: review of wind-tunnel and CFD techniques and their accuracy for wind comfort assessment. Build Environ 100:50–81
Mittal H, Sharma A, Gairola A (2018) A review on the study of urban wind at the pedestrian level around buildings. J Build Eng 18:154–163
Fernando S, Fernando S, Mendis P (2020) Pedestrian wind comfort study using computational fluid dynamic (CFD) simulation. In: Dissanayake R, Mendis P (eds) ICSBE 2018. Lecture notes in civil engineering, vol 44. Springer, Singapore
Zhong J, Liu J, Zhao Y, Niu J, Carmeliet J (2022) Recent advances in modeling turbulent wind flow at pedestrian-level in the built environment. Archit Intell 1:5
Weerasuriya AU, Hu ZZ, Zhang XL, Tse KT, Li S, Chan PW (2018) New inflow boundary conditions for modeling twisted wind profiles in CFD simulation for evaluating the pedestrian-level wind field near an isolated building. Build Environ 132:303–318
Dhunny AZ, Samkhaniani N, Lollchund MR, Rughooputh SDDV (2018) Investigation of multi-level wind flow characteristics and pedestrian comfort in a tropical city. Urban Clim 24:185–204
Cindori M, Čajić P, Džijan I, Juretić F, Kozmar H (2022) A comparison of major steady RANS approaches to engineering ABL simulations. J Wind Eng Ind Aerodyn 221:104867
Tsichritzis L, Nikolopoulou M (2019) The effect of building height and façade area ratio on pedestrian wind comfort of London. J Wind Eng Ind Aerodyn 191:63–75
Hågbo TO, Teigen KEG, Hjertager BH (2021) Influence of geometry acquisition method on pedestrian wind simulations. J Wind Eng Ind Aerodyn 215:104665
Kaseb Z, Hafezi M, Tahbaz M, Delfani S (2020) A framework for pedestrian-level wind conditions improvement in urban areas: CFD simulation and optimization. Build Environ 184:107191
Kikumoto H, Ooka R, Han M, Nakajima K (2018) Consistency of mean wind speed in pedestrian wind environment analyses: mathematical consideration and a case study using large-eddy simulation. J Wind Eng Ind Aerodyn 173:91–99
Ikegaya N, Kawaminami T, Okaze T, Hagishima A (2020) Evaluation of exceeding wind speed at a pedestrian level around a 1:1:2 isolated block model. J Wind Eng Ind Aerodyn 201:104193
Wang W, Okaze T (2022) Estimating low-occurrence wind speeds from mean velocity and turbulent kinetic energy: development of statistical method and validation with idealized cases. Build Environ 224:109499
Okaze T, Kikumoto H, Ono H, Imano M, Ikegaya N, Hasama T, Nakao K, Kishida T, Tabata Y, Yoshie R, Tominaga Y (2017) Large-eddy simulation of flow around buildings: validation and sensitivity analysis. In: proceedings of the 9th Asia-Pacific conference on wind engineering (APCWE), 3–8 December 2017, Auckland, New Zeland
Lin Q, Ishida Y, Tanaka H, Mochida A, Yang Q, Tamura Y (2023) Large eddy simulations of strong wind mechanisms at pedestrian level around square-section buildings with same aspect ratios and different sizes. Build Environ 243:110680
Tamura Y, Xu X, Yang Q (2019) Characteristics of pedestrian-level mean wind speed around square buildings: effects of height, width, size and approaching flow profile. J Wind Eng Ind Aerodyn 192:74–87
Buccolieri R, Santiago JL, Rivas E, Sanchez B (2018) Review on urban tree modelling in CFD simulations: aerodynamic, deposition and thermal effects. Urban For Urban Green 37:56–64
Ricci A, Freda A, Repetto MP, Burlando M, Blocken B (2017) Urban comfort evaluation in an Italian historical district: the impact of architectural details in wind tunnel and CFD analysis. In: Proceedings of the 1st international conference on urban comfort and environmental quality (URBANCEQ-2017), 28–29 September 2017, Genova, Italy
Ramponi R, Le V, Haskell J, Brooks A (2021) Impact of landscaping elements on pedestrian wind simulations using OpenFOAM. In: Proceedings of building simulation 2021: 17th conference of IBPSA, 1–3 Setpember 2021, Bruges, Belgium
Zhong J, Liu J, Xu Y, Liang G (2022) Pedestrian-level gust wind flow and comfort around a building array–influencing assessment on the pocket park. Sustain Cities Soc 83:103953
Fernando HJS, Zajiic D, Di Sabatino S, Dimitrova R, Hedquist B, Dallman A (2010) Flow, turbulence and pollutant dispersion in urban atmospheres. Phys Fluids 22:051301-1–51320
Moonen P, Defraeye T, Dorer V, Blocken B, Carmeliet J (2012) Urban physics: effect of the micro-climate on comfort, health and energy demand. Front Archit Res 1:197–228
Yang S, Wang L, Stathopoulos T, Marey A (2023) Urban microclimate and its impact on built environment—a review. Build Environ 238:110334
Toparlar Y, Blocken B, Maiheu B, van Heijst GJF (2017) A review on the CFD analysis of urban microclimate. Renew Sustain Energy Rev 80:1613–1640
Santamouris M, Osmond P (2020) Increasing green infrastructure in cities: impact of ambient temperature, air quality and heat-releated mortality and morbidity. Buildings 10:233
Lai D, Liu W, Gan T, Liu K, Chen Q (2019) A review of mitigating strategies to improve the thermal environment and thermal comfort in urban outdoor spaces. Sci Total Environ 661:337–353
Ampatzidis P, Cintolesi C, Petronio A, Di Sabatino S, Kershaw T (2022) Evaporating waterbody effects in a simplified urban neighbourhood: a RANS analsys. J Wind Eng Ind Aerodyn 227:105078
Palusci O, Cecere C (2022) Urban ventilation in the compact city: a critical review and a multidisciplinary methodology for improving sustainability and resilience in urban areas. Sustainability 14(7):3948
Bruse M (2004) ENVI-Met 3.0, updated model overview, pp 1–12
Allegrini J, Dorer V, Carmeliet J (2015) Influence of morphologies on the microclimate in urban neighbourhoods. J Wind Eng Ind Aerodyn 144:108–117
Allegrini J, Carmeliet J (2017) Coupled CFD and building energy simulations for studying the impacts of building height topology and buoyancy on local urban microclimate. Urban Clim 21:278–305
Kämpf J (2009) On the modelling and optimisation of urban energy fluxes. Ph.D. thesis, no 4548. EPF, Lausanne
Allegrini J, Carmeliet J (2018) Simulations of local heat islands in Zürich with coupled CFD and building energy models. Urban Clim 24:340–359
Mokhatar S, Reinhart C (2023) Towards scalable and actionable pedestrian outdoor thermal comfort estimation: a progressive modelling approach. Build Environ 242:110547
Kastner P, Dogan T (2022) Eddy3D: a toolkit for decoupled outdoor thermal comfort simulations in urban areas. Build Environ 212:108639
Heidarinejad M, Gracik S, Roudsari MS, Khoshdel Nikkho S, Liu J, Liu K, Pitchorov G, Srebric J (2016) Influence of building surface solar irradiance on environmental temperatures in urban neighborhoods. Sustain Cities Soc 26:186–202
King MF, Gough HL, Halios C, Barlow JF, Robertson A, Hoxey R, Noakes CJ (2017) Investigating the influence of neighbouring structures on natural ventilation potential of a full-scale cubical building using time-dependent CFD. J Wind Eng Ind Aerodyn 169:265–279
Miguel M, Hien WN, Marcel I, Jun Chung HD, Yueer H, Zhonqi Y, Deng JY, Vijayaraghavan SV, Son NN (2021) A physically-based model of interactions between a building and its outdoor conditions at the urban microscale. Energy Build 237:110788
World Health Organization. Ambient (outdoor) air pollution. https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health. Accessed 28 Feb 2024
United Nations (2014) World urbanization prospects: the 2014 revision. United Nations, New York
Nazir MS, Mahdi AJ, Bilal MK, Sohail HM, Ali N, Iqbal HMN (2019) Environmental impact and pollution-related challenges of renewable wind energy paradigm—a review. Sci Total Environ 683:436–444
Li Z, Ming T, Liu S, Peng C, de Richter R, Li W, Zhang H, Wen CY (2021) Review on pollutant dispersion in urban areas—part A: effects of mechanical factors and urban morphology. Build Environ 190:107534
Tominaga Y, Stathopoulos T (2013) CFD simulation of near-field pollutant dispersion in the urban environment: a review of current modeling techniques. Atmos Environ 79:716–730
Qin P, Ricci A, Blocken B (2024) On the accuracy of idealized sources in CFD simulations of pollutant dispersion in an urban street canyon. Buid Environ (paper under review)
Cheng WC, Liu CH (2011) Large-eddy simulation of turbulent transports in urban street canyons in different thermal stabilities. J Wind Eng Ind Aerodyn 99:434–442
Miao Y, Liu S, Zheng Y, Wang S, Liu Z, Zhang B (2015) Numerical study of the effects of planetary boundary layer structure on the pollutant dispersion within built-up areas. J Environ Sci 32:168–179
Mei SJ, Zhao Y, Talwar T, Carmeliet J, Yuan C (2023) Neighborhood scale traffic pollutant dispersion subject to different wind-buoyancy ratios: a LES case study in Singapore. Build Environ 228:109831
Flores F, Garreaud R, Muñoz RC (2014) OpenFOAM applied to the CFD simulation of the turbulent buoyant atmospheric flows and pollutant dispersion inside large open pit mines under intense insolation. Comput Fluids 90:72–87
Reiminger N, Jurando X, Vazquez J, Wemmert C, Blond N, Dufresne M, Wertel J (2020) Effects of wind speed and atmospheric stability on the air pollution reduction rate induced by noise barriers. J Wind Eng Ind Aerodyn 200:104160
Yakhot V, Orszag SA (1986) Renormalization group analysis of turbulence. J Sci Comput 1(1986):3–51
Li W, He Y, Zhang Y, Su J, Chen C, Yu CWF, Zhang R, Gu Z (2019) LES simulation of flow field and pollutant dispersion in a street canyon T under time-varying inflows with TimeVarying-SIMPLE approach. Build Environ 157:185–196
Hang J, Liang J, Wang X, Zhang X, Wu L, Shao M (2022) Investigation of O3–NOx–VOCs chemistry and pollutant dispersion in street canyons with various aspect ratios by CFD simulations. Build Environ 226:109667
Leng S, Li S, Hu ZZ, Wu HY, Li BB (2022) Development of a micro-in-meso-scale framework for simulating pollutant dispersion and wind environment in building groups. J Clear Prod 364:132661
Shao X, Zhao Z, Liu Z, Yang H, Hu C (2023) Exposure risk and emergency evacuation modeling of toxic gas leakage in urban areas under the influence of multiple meteological factors. Environ Pollut 333:122044
Fiates J, Vianna SSV (2016) Numerical modelling of gas dispersion using OpenFOAM. Process Saf Environ Prot 104:266–293
Toja-Silva F, Chen J, Hachinger S, Hase F (2017) CFD simulation of CO2 dispersion from urban thermal power plant: analysis of turbulent Schmidt number and comparison with Gaussian plume model and measurements. J Wind Eng Ind Aerodyn 169:177–193
Zhong J, Cai XM, Bloss WJ (2017) Large eddy simulation of reactive pollutants in a deep urban street canyon: coupling dynamics with O3-NOx-VOC chemistry. Environ Pollut 224:171–184
Janhäll S (2015) Review on urban vegetation and particle air pollution—deposition and dispersion. Atmos Environ 105:130–137
Sanz C (2003) A note on k–ε modelling of vegetation canopy air-flows. Bound Layer Meteorol 108:191–197
Vranckx S, Vos P, Maiheu B, Janssen S (2015) Impact of trees on pollutant dispersion in street canyons: a numerical study of the annual average effects in Antwerp, Belgium. Sci Total Environ 532:474–483
Jeanjean APR, Hinchliffe G, McMullan WA, Monks PS, Leigh RJ (2015) A CFD study on the effectiveness of trees to disperse road traffic emissions at a city scale. Atmos Environ 120:1–14
Jeanjean APR, Monks PS, Leigh RJ (2016) Modelling the effectiveness or urban trees and grass on PM2.5 reduction via dispersion and deposition at a city scale. Atmos Environ 147:1–10
Jeanjean APR, Gallagher J, Monks PS, Leigh RJ (2017) Ranking current and prospective NO2 pollution mitigation strategies: an environmental and economic modelling investigation in Oxford Street, London. Environ Pollut 225:587–597
Wang L, Su J, Gu Z, Tang L (2021) Numerical study on flow field and pollutant dispersion in an ideal street canyon within a real tree model at different wind velocities. Comput Math Appl 81:679–692
McMullan WA, Angelino M (2022) The effect of tree planting on traffic pollutant dispersion in an urban street canyon using large eddy simulation with a recycling and rescaling inflow generation method. J Wind Eng Ind Aerodyn 221:104877
Derome D, Kulibaly A, Defraeye T, Blocken B, Carmeliet J (2017) Ten questions concerning modeling of wind-driven rain in the built environment. Build Environ 114:495–506
Blocken B, Carmeliet J (2004) A review of wind-driven rain research in building science. J Wind Eng Ind Aerodyn 92:1079–1130
Blocken B, Carmeliet J (2010) Overview of the three state-of-the-art wind driven rain assessment models and comparison based on model theory. Build Environ 45(3):691–703
Gholamalipour P, Ge H, Stathopoulos T (2022) Wind-driven rain (WDR) loading on building facades: a state-of-the-art review. Build Environ 221:1093314
Kulibaly A, Derome D, Blocken B, Carmeliet J (2014) Numerical simulations of wind-driven rain on an array of low-rise cubic buildings and validation by field measurements. Build Environ 81:283–295
Kulibaly A, Derome D, Carmeliet J (2017) Analysis of time-resolved wind-driven rain on an array of low-rise cubic buildings using large eddy simulation and an Eulerian multiphase model. Build Environ 114:68–81
Kulibaly A, Derome D, Blocken B, Carmeliet J (2015) Wind-driven rain on two parallel wide buildings: field measurements and CFD simulations. J Wind Eng Ind Aerodyn 146:11–28
Chen H, Zhou B (2022) Numerical simulation of wind-driven rain based on the Eulerian model. Front Earth Sci 10:917514
Xu G, Tan K, Ge Z, Poh HJ, Ooi C, Eng Y (2023) Automatic selection of release plane for Lagrangian-based wind-driven rain studies. J Wind Eng Ind Aerodyn 232:105242
Bruno L, Horvat M, Raffaele L (2018) Windblown sand along railway infrastructures: a review of challenges and mitigation measures. J Wind Eng Ind Aerodyn 177:340–365
Smyth TAG (2016) A review of computational fluid dynamics (CFD) airflow modelling over aeolian landforms. Aeolian Res 22:153–164
Deaves DM (1975) Wind over hills—a numerical approach. J Ind Aerodyn 1:371–391
Deaves DM (1980) Computations of wind flow over two-dimensional hills and embankments. J Ind Aerodyn 6:89–111
Zhang J, Cui E, Fu G (1995) Investigation of the flow field and the starting conditions of wind-induced erosion of the railway embankemnt. J Wind Eng Ind Aerodyn 54–55:573–581
Bruno L, Fransos D (2015) Sand transverse dune aerodynamics: 3D coherent flow structures from a computational study. J Wind Eng Ind Aerodyn 147:291–301
Menter FR (1997) Eddy viscosity transport equations and their relation to the k–ε model. J Fluid Eng 119:876–884
Bruno L, Fransos D, Lo Giudice A (2018) Solid barriers for windblown sand mitigation: aerodynamic behavior and conceptual design guidelines. J Wind Eng Ind Aerodyn 173:79–90
Horvat M, Bruno L, Khris S (2021) CWE study of wind flow around railways: effects of embankment and track system on sand sedimentation. J Wind Eng Ind Aerodyn 208:104476
Pourteimouri P, Campmans GHP, Wijnberg KM, Hulscher SJMH (2022) A numerical study on the impact of building dimensions on airflow patterns and bed morphology around buildings at the beach. J Mar Sci Eng 10:13
Pourteimouri P, Campmans GHP, Wijnberg KM, Hulscher SJMH (2023) How wind direction and building spacing influences airflow patterns and sediment transport patterns around a row of beach buildings: a numerical study. Aeolian Res 61:100867
Hesp P, Smyth TAG (2019) CFD flow dynamics over model scarps and slops. Phys Geogr 42(1):1–24
Tominaga Y, Okaze T, Mochida A (2011) CFD modeling of snowdrift around a building: an overview of models and evaluation of a new approach. Build Environ 46(4):899–910
Tominaga Y (2018) Computational fluid dynamics simulation of snowdrift around buildings: past achievements and future perspectives. Cold Reg Sci Technol 150:2–14
Uematsu T, Nakata T, Takeuchi K, Arisawa Y, Kaneda Y (1991) Three-dimensional numerical simulation of snowdrift. Cold Reg Sci Technol 20:65–73
Bang B, Nielsen A, Sundsbo PA, Wiik T (1994) Computer simulation of wind speed, wind pressure and snow accumulation around buildings (SNOW-SIM). Energy Build 21:235–243
Sekine A, Shimura M, Maruoka A, Hirano H (1999) The numerical simulation of snowdrift around a building. Int J Comput Fluid Dyn 12(3–4):249–255
Tominaga Y, Mochida A, Yoshino H (2002) Numerical analysis of wind environment and snowdrift around building complex based on CFD and CG techniques. J Snow Eng 18(1):3–11 (in Japanese)
Hames O, Jafari M, Wagner DN, Raphael I, Clemens-Sewall D, Polashenski C, Shupe MD, Schneebeli M, Lehning M (2021) Modelling the small-scale deposition of snow onto structured Arctic sea ice during a MOSAiC storm using snowBedFoam 1.0. Geosci Model Dev 15:6429–6449
Chen X, Yu Z (2022) DriftScalarDyFoam: an OpenFOAM-based multistage solver for drifting snow and its distribution around buildings. Front Earth Sci 10:822140
Chen X, Yu Z (2023) Development of Eulerian–Lagrangian simulation for snow transport in the presence of obstacles. Cold Reg Sci Technol 206:103730
Lü X, Huang N, Tong D (2012) Wind tunnel experiments on natural snow drift. Sci China Technol Sci 55:927–938
Lo Giudice A, Preziosi L (2020) A fully Eulerian multiphase model of windblown sand coupled with morphodynamic evolution: erosion, transport, deposition, and avalanching. Appl Math Model 79:68–84
Bruno L, Salvetti MV, Ricciardelli F (2014) Benchmark on the aerodynamics of a rectangular 5:1 cylinder: an overview after the first four years of activity. J Wind Eng Ind Aerodyn 126:87–106
Nieto F, Owen JS, Hargreaves DM, Hernández S (2015) Bridge deck flutter derivatives: efficient numerical evaluation exploiting theri interdependence. J Wind Eng Ind Aerodyn 136:138–150
Šarkic Glumac A, Höffer R, Brčić S (2015) Numerical simulations and experimental validations of force coefficients and flutter derivatives of a bridge deck. J Wind Eng Ind Aerodyn 144:172–182
Andersen MS, Bossen NS (2021) Modal decomposition of the pressure field on a bridge deck under vortex shedding using POD, DMD and ERA with correlation functions as Markov parameters. J Wind Eng Ind Aerodyn 215:104699
Li K, Ge YJ, Zhao L, Guo ZW (2017) Numerical simulation of feedback control for a single-box-girder suspension bridge by twin-winglet system. J Wind Eng Ind Aerodyn 169:77–93
Xu M, Patruno L, Lo YL, de Miranda S (2022) Simulation strategies for wind shields and porous barriers for bridge deck optimization. Structures 40:824–839
Álvarez AJ, Nieto F, Kwok KCS, Hernández S (2018) Aerodynamic performance of the twin-box decks: a parametric study gap width effects based on validated 2D URANS simulations. J Wind Eng Ind Aerodyn 182:202–221
Nieto F, Cid Montoya M, Hernández S, Kusano I, Casteleiro A, Álvarez AJ, Jurado AA, Fontán A (2020) Aerodynamic and aerolastic responses of short gap twin-box decks: Box geometry and gap distance dependent surrogate based design. J Wind Eng Ind Aerodyn 201:104147
Álvarez AJ, Nieto F, Kwok KCS, Hernández S (2021) A computational study on the aerodynamics of a twin-box bridge with a focus on the spanwise features. J Wind Eng Ind Aerodyn 209:104465
Zhu L, McCrum D, Sweeney C, Keenahan J (2023) Full-scale computational fluid dynamics study on wind condition of the long-span Queensferry Crossing Bridge. J Civ Struct Health Monit 13:615–632
Moore N, Keenahan J (2022) Wind forces on medium-span bridges: a comparison of Eurocode 1 Part 4 and computational fluid dynamics. Civ Eng 3:794–808
Zhang Y, MacReamoinn R, Cardiff P, Keenahan J (2023) Analyzing wind effects on long-span bridges: a viable numerical modelling methodology using OpenFOAM for industrial applications. Infrastructures 8:130
Hao J, Wu T (2018) Downburst-induced transient response of a long-span bridge: a CFD-CSD-based hybrid approach. J Wind Eng Ind Aerodyn 179:273–286
Qin Y, Cao J, Cao S, Ge Y (2022) Numerical simulation of wind loads and aerodynamic characteristics of streamlined bridge decks under tornado-like vortices. J Fluids Struct 113:103623
Chen Y, Wan X, Sun C, Zhu B (2024) Exploring the failure mechanism of light poles on elevated bridges under high winds. Eng Fail Anal 159:108076
Wardlaw RL, Moss GF (1970) A standard tall building model for the comparison of simulated natural winds in wind tunnels, C.A.A.R.C., C.C.662m Tech.25
Mochida A, Murakami S, Shoji M, Ishida Y (1993) Numerical simulation of flowfield around texas tech building by large eddy. SIMULATION 46–47:455–460
Stathopoulos T, Zhou YS (1993) Numerical simulation of wind-induced pressures on buildings of various geometries. J Wind Eng Ind Aerodyn 46–47:419–430
Huang MF, Lau IWH, Chan CM, Kwok KCS, Li G (2011) A hybrid RANS and kinematic simulation of wind load effects on full-scale tall buildings. J Wind Eng Ind Aerodyn 99:1126–1138
Joubert EC, Harms TM, Venter G (2015) Computational simulation of the turbulent flow around a surface mounted rectangular prism. J Wind Eng Ind Aerodyn 142:173–187
Wang Y, Chen X (2020) Simulation of approaching boundary layer flow and wind loads on high-rise buildings by wall-modeled LES. J Wind Eng Ind Aerodyn 207:104410
Wang Y, Chen X (2022) Evaluation of wind loads on high-rise buildings at various angles of attack by wall-modelled large-eddy simulation. J Wind Eng Ind Aerodyn 229:105160
Wei D, Spence SMJ, Kareem A, Jemcov A (2014) A structured mesh boundary motion approach for simulating wind effects on bluff bodies with changing boundaries. J Wind Eng Ind Aerodyn 126:118–131
Agrawal S, Wong JK, Song J, Mercan O, Kushner PJ (2022) Assessment of the aerodynamic performance of unconventional building shapes using 3D steady RANS with SST k–ω turbulence model. J Wind Eng Ind Aerodyn 225:104988
Villanueva L, Valero MM, Šarkic Glumac A, Meldi M (2024) Augmented state estimation of urban settings using on-the-fly sequential data assimilation. Comput Fluids 269:106118
Gimenez JM, Bre F (2019) Optimization of RANS turbulence models using genetic algorithms to improve the prediction of wind pressure coefficients on low-rise buildings. J Wind Eng Ind Aerodyn 193:103978
Lamberti G, Gorlé C (2021) A multi-fidelity machine learning framework to predict wind loads on buildings. J Wind Eng Ind Aerodyn 214:104647
Šarkic Glumac A, Jadhav O, Despotović V, Blocken B (2023) A multi-fidelity wind surface pressure assessment via machine learning: a high-rise building case. Build Environ 234:110135
Kim Y, Castro IP, Xie ZT (2013) Divergence-free turbulence model inflow conditions for large-eddy simulations with incompressible flow solvers. Comput Fluids 84:56–68
Yan BW, Li QS (2015) Inflow turbulence generation methods with large eddy simulation for wind effects on tall buildings. Comput Fluids 116:158–175
Ricci M, Patruno L, Kalkman I, de Miranda S, Blocken B (2018) Towards LES as a design tool: wind loads assessment on a high-rise building. J Wind Eng Ind Aerodyn 180:1–18
Lamberti G, Gorlé C (2020) Sensitivity of LES predictions of wind loading on a high-rise building to the inflow boundary condition. J Wind Eng Ind Aerodyn 206:104370
Daniels SJ, Xie ZT (2022) Overview of large-eddy simulation for wind loading on slender structures. Proc Inst Civ Eng Eng Comput Mech 175(2):41–71
Geleta TN, Bitsuamlak G (2022) Validation metrics and turbulence frequency limits for LES-based wind load evaluation for low-rise buildings. J Wind Eng Ind Aerodyn 231:105210
Potsis T, Stathopoulos T (2024) Wind induced peak pressures on low-rise building roofs via dynamic terrain computational methodology. J Wind Eng Ind Aerodyn 245:105630
Zhang Y, Cao S, Cao J (2022) An improved consistent inflow turbulence generator for LES evaluation of wind effects on buildings. Build Environ 223:109459
Melaku AF, Bitsuamlak GT (2024) Prospect of LES for predicting wind loads and responsed of tall buildings: a validation study. J Wind Eng Ind Aerodyn 244:105613
Xing J, Patruno L, Pozzuoli C, Pedro G, de Miranda S, Ubertini F (2022) Wind loads prediction using LES: inflow generation, accuracy and cost assessment for the case of Torre Gioia 22. Eng Struct 262:114292
Marsland L, Nguyen K, Zhang Y, Huang Y, Abu-Zidan Y, Gunawardena T, Mendis P (2022) Improving aerodynamic performance of tall buildings using facade openings at service floors. J Wind Eng Ind Aerodyn 225:104997
Hasama T, Saka T, Itoh Y, Kondo K, Yamamoto M, Tamura T, Yokokawa M (2020) Evaluation of aerodynamic instability for building using fluid–structure interaction analysis combined with multi-degree-of-freedom structure model and large-eddy simulation. J Wind Eng Ind Aerodyn 197:104052
Liu J, Hui Y, Yang Q, Zhang R (2023) LES evaluation of the aerodynamic characteristics of high-rise building with horizontal ribs under atmospheric boundary layer flow. J Build Eng 71:106487
Melaku AF, Doddipatla LS, Bitsuamlak GT (2022) Large-eddy simulation of wind loads on a roof-mounted cube: application for interpolation of experimental aerodynamic data. J Wind Eng Ind Aerodyn 232:105230
Sim TS, Ong MC, Quek WY, Sum ZW, Lai WX, Skote M (2016) A numerical study of microburst-like wind load acting on different block array configurations using an impinging jet model. J Fluids Struct 61:184–204
Žužul J, Ricci A, Burlando M, Blocken B, Solari G (2023) CFD analysis of the WindEEE dome produced downburst-like winds. J Wind Eng Ind Aerodyn 232:105268
Menter FR, Egorov Y (2010) The scale-adaptive simulation method for unsteady turbulent flow predictions. Part 1: theory and model description. Flow Turbul Combust 85:113–138
Iida Y, Uematsu Y (2019) Numerical study of wind loads on buildings induced by downbursts. J Wind Eng Ind Aerodyn 191:103–116
Cao S, Wang M, Cao J (2018) Numerical study of wind pressure on low-rise buildings induced by tornado-like flows. J Wind Eng Ind Aerodyn 183:214–222
Roy S, Kundu CK (2021) State of the art review of wind induced vibration and its control on transmission towers. Structures 29:254–264
Holmes JD (1994) Along-wind response of lattice towers: part 1—derivation of expressions for gust response factors. Eng Struct 16(4):287–292
Holmes JD (1996) Along-wind response of lattice towers: part 2—aerodynamic damping and deflections. Eng Struct 18(7):483–488
Holmes JD (1996) Along-wind response of lattice towers: part 3—effective load distributions. Eng Struct 18(7):489–494
Aboshosha H, Elawady A, El Ansary A, El Damatty AA (2016) Review on dynamic and quasi-static buffeting response of transmission lines under synoptic and non-synoptic winds. Eng Struct 112:23–46
Fabre S, Stickland M, Scanlon T, Odroyd A, Kindler D, Quail F (2014) Measurement and simulation of the flow field around the FINO 3 triangular lattice meteorological mast. J Wind Eng Ind Aerodyn 130:99–107
Zhang D, Hu X, Song X, Deng H, Ma X (2023) Investigation on aerodynamic characteristics for steel tubular cross-arms of transmission tower under skew wind. J Wind Eng Ind Aerodyn 222:104914
Hadane A, Redford JA, Gueguin M, Hafid F, Ghidaglia JM (2023) CFD wind tunnel investigation for wind loading on angle members in lattice tower structures. J Wind Eng Ind Aerodyn 236:105397
Ma T, Sun C (2021) Large eddy simulation of hurricane boundary layer turbulence and its application for power transmission system. J Wind Eng Ind Aerodyn 210:104520
Nastasi B, Markovska N, Puksec T, Duić N, Foley A (2022) Renewable and sustainable energy challenges to face for the achievement of sustainable development goals. Renew Sustain Energy Rev 157:112071
Zhou Q, Dong P, Li M, Wang Z (2023) Analyzing the interactions between photovoltaic system and its ambient environment using CFD techniques: a review. Energy Build 296:113394
Jubayer CM, Hangan H (2014) Numerical simulation of wind effects on a stand-alone ground mounted photovoltaic (PV) system. J Wind Eng Ind Aerodyn 134:56–64
Jubayer CM, Hangan H (2016) Numerical approach to the investigation of wind loading on an array of ground mounted solar photovoltaic (PV) panels. J Wind Eng Ind Aerodyn 153:60–70
Sun J, He Y, Li X, Lu Z, Yand X (2023) CFD simulations for layout optimal design for ground-mounted photovoltaic panel arrays. J Wind Eng Ind Aerodyn 242:105558
Wang J, Yang Q, Van Phuc P, Tamura Y (2020) Characteristics of conical vortices and their effects on wind pressures on flat-roof-mounted solar arrays by LES. J Wind Eng Ind Aerodyn 200:104146
Aly AM, Clarke J (2023) Wind design of solar panels for resilient and green communities: CFD with machine learning. Sustain Cities Soc 94:104529
OpenFOAM Foundation. https://openfoam.org/. Accessed 02 May 2024
OpenFOAM ESI. https://openfoam.com. Accessed 02 May 2024
FOAM extend. https://sourceforge.net/projects/foam-extend/. Accessed 02 May 2024
Solari G, Burlando M, Repetto MP (2020) Detection, simulation, modelling and loading of thunderstorm outflows to design wind-safer and cost-efficient structures. J Wind Eng Ind Aerodyn 200:104142
THUNDERR (2024) http://www.thunderr.eu/projects/. Accessed 02 May 2024
OpenFOAM journal. https://journal.openfoam.com/index.php/ofj. Accessed 02 May 2024
Acknowledgements
The author gratefully acknowledges the organizers of the 18th OpenFOAM workshop, Prof. J. O. Pralits and Prof. J. Guerrero, for the kind invitation to organize in cooperation with Prof. M. Burlando the technical session "Civil Engineering and Wind Engineering" and to draft this review paper on the use of OpenFOAM in the Wind Engineering. The author is also very grateful to Prof. M. Burlando for his support during the drafting of this review paper. The author expresses his gratitude to the Editor and Reviewers for providing very valuable comments and suggestions to further improve the paper.
Funding
No funds have been received for this study.
Author information
Authors and Affiliations
Contributions
Conceptualization, Formal analysis, Data curation, Writing, Editing, Revising
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Conflict of interest
The author declares that there is no conflict of interest.
Ethical approval
Not applicable.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Ricci, A. Review of OpenFOAM applications in the computational wind engineering: from wind environment to wind structural engineering. Meccanica (2024). https://doi.org/10.1007/s11012-024-01826-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11012-024-01826-x