Abstract
There is a growing need for accurate and interpretable machine learning models of thermal comfort in buildings. Physics-informed machine learning could address this need by adding physical consistency to such models. This paper presents metamodeling of thermal comfort in non-air-conditioned buildings using physics-informed machine learning. The studied metamodel incorporated knowledge of both quasi-steady-state heat transfer and dynamic simulation results. Adaptive thermal comfort in an office located in cold and hot European climates was studied with the number of overheating hours as index. A one-at-a-time method was used to gain knowledge from dynamic simulation with TRNSYS software. This knowledge was used to filter the training data and to choose probability distributions for metamodel forms alternative to polynomial. The response of the dynamic model was positively skewed; and thus, the symmetric logistic and hyperbolic secant distributions were inappropriate and outperformed by positively skewed distributions. Incorporating physical knowledge into the metamodel was much more effective than doubling the size of the training sample. The highly flexible Kumaraswamy distribution provided the best performance with R2 equal to 0.9994 for the cold climate and 0.9975 for the hot climate. Physics-informed machine learning could combine the strength of both physics and machine learning models, and could therefore support building design with flexible, accurate and interpretable metamodels.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Beucler T, Rasp S, Pritchard M, et al. (2019). Achieving conservation of energy in neural network emulators for climate modeling. arXiv:1906.06622
Breesch H, Janssens A (2010). Performance evaluation of passive cooling in office buildings based on uncertainty and sensitivity analysis. Solar Energy, 84: 1453–1467.
Bünning F, Huber B, Schalbetter A, et al. (2022). Physics-informed linear regression is competitive with two Machine Learning methods in residential building MPC. Applied Energy, 310: 118491.
CEN (2019). EN 16798-1: Energy performance of buildings — ventilation for buildings — Part 1: Indoor environmental input parameters for design and assessment of energy performance of buildings addressing indoor air quality, thermal environment, lighting and acoustics. Brussels: European Committee for Standardization.
Chen X, Yang H, Sun K (2017). Developing a meta-model for sensitivity analyses and prediction of building performance for passively designed high-rise residential buildings. Applied Energy, 194: 422–439.
de Dear R, Brager G (1998). Developing an adaptive model of thermal comfort and preference. ASHRAE Transactions, 104(1): 145–167.
Drgoňa J, Tuor AR, Chandan V, et al. (2021). Physics-constrained deep learning of multi-zone building thermal dynamics. Energy and Buildings, 243: 110992.
Eisenhower B, O’Neill Z, Narayanan S, et al. (2012). A methodology for meta-model based optimization in building energy models. Energy and Buildings, 47: 292–301.
Fletcher MJ, Johnston DK, Glew DW, et al. (2017). An empirical evaluation of temporal overheating in an assisted living Passivhaus dwelling in the UK. Building and Environment, 121: 106–118.
Fréchet M (1928). Sur la loi de probabilité de l’écart maximum. In Annales de la societe Polonaise de Mathematique. (in French)
Haldi F, Robinson D (2008). On the behaviour and adaptation of office occupants. Building and Environment, 43: 2163–2177.
Harlow DG (2002). Applications of the Fréchet distribution function. International Journal of Materials and Product Technology, 17: 482–495.
Jaffal I, Inard C, Ghaddar N, et al. (2020). A metamodel for long-term thermal comfort in non-air-conditioned buildings. Architectural Engineering and Design Management, 16: 441–472.
Jones MC (2009). Kumaraswamy’s distribution: A beta-type distribution with some tractability advantages. Statistical Methodology, 6: 70–81.
Karniadakis GE, Kevrekidis IG, Lu, Perdikaris P, et al. (2021). Physics-informed machine learning. Nature Reviews Physics, 3: 422–440.
Khandelwal A, Xu S, Li X, et al. (2020). Physics guided machine learning methods for hydrology. arXiv:2012.02854.
Klein SA, Beckman WA, Mitchell JW, et al. (2004). TRNSYS 16 — A TRaNsient SYstem Simulation program, user manual. Madison, WI, USA: Solar Energy Laboratory, University of Wisconsin—Madison.
Kotz S, Nadarajah S (2000). Extreme Value Distributions: Theory and Applications. Singapore: World Scientific.
Kumaraswamy P (1980). A generalized probability density function for double-bounded random processes. Journal of Hydrology, 46: 79–88.
Li YF, Ng SH, Xie M, et al. (2010). A systematic comparison of metamodeling techniques for simulation optimization in Decision Support Systems. Applied Soft Computing, 10: 1257–1273.
Li B, Yao R, Wang Q, et al. (2014). An introduction to the Chinese Evaluation Standard for the indoor thermal environment. Energy and Buildings, 82: 27–36.
Lomas KJ, Porritt SM (2017). Overheating in buildings: Lessons from research. Building Research and Information, 45: 1–18.
Magnier L, Haghighat F (2010). Multiobjective optimization of building design using TRNSYS simulations, genetic algorithm, and Artificial Neural Network. Building and Environment, 45: 739–746.
Morgan C, Foster JA, Poston A, et al. (2017). Overheating in Scotland: Contributing factors in occupied homes. Building Research and Information, 45: 143–156.
Nagler J (1994). Scobit: an alternative estimator to logit and probit. American Journal of Political Science, 38: 230–255.
Nicol JF, Humphreys MA (2002). Adaptive thermal comfort and sustainable thermal standards for buildings. Energy and Buildings, 34: 563–572.
Rackes A, Melo AP, Lamberts R (2016). Naturally comfortable and sustainable: informed design guidance and performance labeling for passive commercial buildings in hot climates. Applied Energy, 174: 256–274.
Rossi MM, Oliveira Favretto AP, Grassi C, et al. (2019). Metamodels to assess the thermal performance of naturally ventilated, low-cost houses in Brazil. Energy and Buildings, 204: 109457.
Simpson TW, Poplinski JD, Koch PN, et al. (2001). Metamodels for computer-based engineering design: survey and recommendations. Engineering With Computers, 17: 129–150.
Symonds P, Taylor J, Chalabi Z, et al. (2015). Performance of neural networks vs. Radial basis functions when forming a metamodel for residential buildings. International Journal of Civil and Environmental Engineering, 9: 1594–1598.
Takasu M, Ooka R, Rijal HB, et al. (2017). Study on adaptive thermal comfort in Japanese offices under various operation modes. Building and Environment, 118: 273–288.
van Gelder L, Das P, Janssen H, et al. (2014). Comparative study of metamodelling techniques in building energy simulation: Guidelines for practitioners. Simulation Modelling Practice and Theory, 49: 245–257.
von Rueden L, Mayer S, Garcke J, et al. (2019). Informed machine learning—towards a taxonomy of explicit integration of knowledge into machine learning. Learning, 18: 19–20.
Zobeiry N, Humfeld KD (2021). A physics-informed machine learning approach for solving heat transfer equation in advanced manufacturing and engineering applications. Engineering Applications of Artificial Intelligence, 101: 104232.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The authors have no competing interests to declare that are relevant to the content of this article.
Rights and permissions
About this article
Cite this article
Jaffal, I. Physics-informed machine learning for metamodeling thermal comfort in non-air-conditioned buildings. Build. Simul. 16, 299–316 (2023). https://doi.org/10.1007/s12273-022-0931-y
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12273-022-0931-y