Effect analysis of U-shape exterior walls on energy consumption of building: the case of Morocco

The building architecture significantly affects the energy consumption of buildings. In this paper, we study the effect of using U-shape exterior walls on energy consumption. The main target is to evaluate the impact of this parameter on heating and cooling loads for a small building model in Tetouan, Morocco (an administrative building divided into 2 zones, the effective area is 27 m2). In this context, a parametric study was carried out by the TRNSYS software 18, in order to evaluate the efficiency of using this form of exterior walls according to two selected criteria: the orientation of the exterior wall and the depth of the U-shape. More precisely, five values of the U-shape depth were studied for each orientation (South, North, South-east, and South-west), in six climate zones in Morocco presenting different climate conditions. In this sense, 126 simulations were done to have as a result the heating and cooling load for each scenario. The results showed a significant difference in the total load of the building model using different depths of the U-shape exterior walls in different orientations. We found that the U-shape parameter modified in the exterior walls is more efficient in zones characterized by a cold to moderate climate, namely, Ifran, Rabat, and Tangier (the maximum reduction of heating and cooling demand varies from 3.6 to 14% depending on the climate zone). Also, a maximum reduction in the total heating and cooling consumption is noticeable in zones with a hot climate which fluctuates between 1.9 and 3.1%.


Introduction
The energy consumption in buildings is considered as an important part of the world's energy consumption and greenhouse gas emissions. Therefore, reducing the building's energy consumption and optimizing its energy performance are two key-target areas for several researchers [1][2][3][4]. Conventionally, previous studies have used different approaches based on the evaluation of several alternative design options to identify the best solution. Al-Saggaf et al. [5] studied the impact of architectural design features (building orientation; building envelope) on three different designs to evaluate heating, ventilation, and air conditioning (HVAC) loads excessive consumption in hot climatic conditions. The numerical simulation results showed that design C is the best design alternative in terms of energy performance with a cost savings part of about 27% compared to designs A and B. Azmi et al.
[6] conducted a comprehensive literature review with the aim of identifying important aspects to improve the thermal performance of mosque building. They discussed the impact of different components of the mosque envelope such as walls, roofs, and windows. Albatayneh [7] presented an optimization study of several parameters of the building envelope (building orientation, thickness of outside wall insulation, thickness of floor and roof insulation, glazing type). The results showed that the consumption of energy was reduced by 94.38%, 91.32%, and 95.24% for the overall energy usage, the cooling load, and the heating load, respectively. Zhu et al.
[8] investigated a new optimization method for building envelope design for the lowest carbon emissions and reducing the environmental impact of building operational energy consumption using orthogonal experimental design (OED). Besides that, Pathirana et al. [9] examined the effect of building shape, zones, window, and orientation on thermal comfort in the naturally ventilated tropical climate. The results indicate that a rectangular shape with a staircase positioned in the middle will provide higher thermal comfort.
A similar research has been carried out to identify the impact of building shape on energy load. Considering the different aspects such as facades with large glazed portions, large surfaces exposed directly to the sun especially in climates with canicular days, shading panels, and interbuilding effects [10-13]. Mirrahimi et al. [14] exposed a detailed review of the most important parameters affecting the energy efficiency for high-rise buildings. This analysis leads to the conclusion that a strong relationship exists between various building components such as external wall, external roof and external glazing, insulation, and the reduction of energy consumption. Liu et al. [15] conducted a developed research about Chinese universities (10 typical universities in northern China). They were investigated in terms of planning and layout, day lighting, ventilation, thermal comfort, and system energy use. It is recommended from this study that a building should combine energy efficiency with the design planning and layout, functional design, construction design, and system design. Besides that, the building orientation has a significant effect on energy consumption. Therefore, it is vital to consider south-east orientation in most climate zones [16,17].
As mentioned earlier, there are several researches in the literature regarding the different envelope parameter effects on building energy consumption. However, there is a lack of studies that deal with the building shape of the external walls. For this purpose, the main objective of the current study is to develop and evaluate the effect of changing the habitual form of exterior walls by U-shape ones for a small building model located in Tetouan (Morocco). Furthermore, the U-shaped buildings are a widely distributed building layout. Their semi-enclosed plane form is not only conducive to ventilation and dehumidification but also relatively compact without excessive heat loss compared to other forms [18,19]. Hence, the heating and cooling loads are calculated for the reference case; then, the same model is used with different scenarios (five scenarios for each orientation) in order to study the effect of the U-shape depth of the wall on energy consumption.

Methodology
The aim of this study is to determine the energy-efficient scenario concerning the shape of exterior walls (using different values of U-shape depth in exterior walls for different orientations) that can be used to increase the energy performance of an office building in the six climate zones of Morocco as shown in Fig. 1 [20]. The Mediterranean climate of North Africa is characterized by hot dry summers and seasonally restricted rainfall. The western region experiences a sub-humid Mediterranean climate with mild, moist winters from October through March/April and hot, dry summers from May to September [21]. For this reason, six cities were chosen to represent the six climate zones as shown in Fig. 1 (Tangier, Ifran, Errachidia, Rabat, Marrakech, and Meknes).
The methodology of the study involves three steps, as can be seen in Fig. 2. The first one is the modeling step. In this step, the dimensions and properties of the hypothetical module, given in Section 2.1, were determined. Furthermore, 126 scenarios (5 scenarios for four orientations in the six climate zones as described in Table 2) were created to be applied to the module in order to evaluate the impact of using different U-shape values on the energy performance of our building model. After obtaining the climate data of the location, all data were transferred to the TRNSYS program, which is used to perform dynamic energy simulations, which is the second step of the methodology. In this step, assumptions and limitations of the model were introduced to the program as detailed in Section 2.3. Then, annual heating and cooling load values for all scenarios were calculated using the TRNSYS energy simulation program (see Fig. 5).

Building model
The simulation concerns a small administrative building, which consists of one level as shown in Fig. 3a [22]. The effective area is 27m 2 , divided into 2 zones. The windowto-wall ratio is 18%. Figure 3b shows the area allocated for the two zones. The composition of the structural elements is presented in Table 1.
The current study is made according to a reference case which consists of considering the basic model (as given in Fig. 3a), without any improvement of the energy efficiency of the building. The total load of heating and cooling is calculated for this model and then compared to other scenarios proposed concerning the replacement of an exterior wall form by a U-shape one with different depth values of the U-shape.
In view to study the impact of this parameter, 126 simulation cases (Table 2) have been proposed (four cases in four orientations for six climate zones) as follows: Scenario 1: Variation of the form of one exterior wall by a U-shape one. The U-shape depth is 0.25 m. Scenario 2: Variation of the form of one exterior wall by a U-shape one. The U-shape depth is 0.5 m. Scenario 3: Variation of the form of one exterior wall by a U-shape one. The U-shape depth is 0.75 m. Scenario 4: Variation of the form of one exterior wall by a U-shape one. The U-shape depth is 1 m. Scenario 5: Variation of the form of one exterior wall by a U-shape one. The U-shape depth is 1.25 m.
These different cases are presented in the figure below for the south orientation (Fig. 4). The five scenarios are studied for the three other facades (south-east, southwest, and north) in order to evaluate the impact of the U-shape wall in all facades of the building model (four simulations for each façade corresponding to four different depths of the U-shape wall).

Simulation settings
The simulations of this study were conducted by the TRNSYS software. Each zone corresponds to one air node and represents the thermal capacity of the zone air volume. In this model, any air point in the zone has the same temperature, humidity, and other properties. The TRNSYS output in the multi-zone building component "NTYPE 904" represents the energy balance for a zone using the following equation: With the following: DQair dt : the change in internal energẏ Qheat : heating demand (convective + radiative) Q COOL : cooling demanḋ Q vent : ventilation gainṡ Q inf : infiltration gainṡ Q trans : transmission gains into the wall Q gain : internal gains (convective and radiative) Q sol : absorbed solar gains on all inside surfaces of zones The thermal simulation was done using Simulation Studio and Type 56 in the TRNSYS software [23,24]. Then, a series of thermal simulations was carried out on the building model using the TRNSYS software.

Assumptions and limitations
To evaluate the efficiency of using the parametric study, the following assumptions and limitations about the building model have been taken into account. (1)

Occupancy
It was assumed that the human body is considered as a thermal generator, whose power depends on the activity exerted. In our case, the number of occupants in our building model is 2 persons/ zone. The occupancy profile is considered as given in Table 3. The TRNSYS software presents several types of gains due to the occupants based on the ISO 7730 norm; we choose according to the type of building the following power: 100 W/person.

Contributions due to electrical appliances and lighting
Computers: electrical power 230 W/post. The administrative building studied has 4 computers.

Ventilation rate
Ventilation rate including infiltration is 0.6 vol/h, according to Moroccan standard NM ISO 13789/2010.

Set temperature
In winter, the heating turns on when the zone temperatures fall below 20 °C, and during summer, the cooling turns on when the temperature exceeds 26 °C.

Model validation and meteorological data.
A dynamic simulation analysis was conducted by the TRNSYS 18 software to determine the effect of the U-shape depth parameter on the heating and cooling loads of the studied administrative building (see Section 2.1). The use of this software has been reported by different researchers [25]. For this purpose, a small building located in Tetouan (Morocco) [22] was used in our study as a model to build a well-validated simulation compared to the results of the total heating and cooling loads of this building. So, the same building model is used to evaluate the impact of the studied parameter on thermal performance and energy savings. It can be clear from the validation model in Table 4 that the relative error of the total heating and cooling load building does not exceed 2.7%. This difference is due to the simulation hypotheses used in this simulation which may be different from those of the validation paper [22], namely the climatic data and the occupancy mode of the building. In fact, to carry out the dynamic thermal simulations, hourly meteorological data (ambient temperature, relative humidity, and the direct and diffuse horizontal radiation) are necessary for the TRNSYS software; these data are imported from the "Metronome" software [26].

Results and discussion
The design and simulation of the current model are shown in Fig. 6. Hence, all the results of the 3D model using different U-shape depths of the exterior wall were developed and analyzed in this section. The simulation results of all climate zones show that the air conditioning load far exceeds (by 1.5 to 6.8 times) the heating load for all climate zones according to the orientation of the building, except for the Ifran zone where the heating load is 3.6 to 4.5 times greater than the air conditioning load depending on the orientation, this difference between cooling and heating loads confirm that the percentages of cooling reduction are more significant than heating ones. By considering the total heating and cooling load, a significant change can be observed according to the U-shape depth of the exterior wall and its orientation (Tables 5, 6, and 7). From the findings, it can be clear that the integration of the U-shape always reduces the total heating and cooling load for the four orientations studied according to the U-shape depth, for all climate zones,    except the south-west orientation in Marrakech that gives no total consumption gain (Fig. 7). Furthermore, for all climate zones, the maximum reduction values for total consumption are observed in the northern and southern orientations. The U-shape parameter modified in the exterior walls is more efficient in zones characterized by a cold to moderate climate, namely, Ifran, Rabat, and Tangier (the total reduction of heating and cooling demand varies from 3.6 to 14%), while warm zones (Meknes, Marrakech, and Errachidia) present a smaller reduction of the total load (from 1.9 to 3.1%). Then, for all climate zones, the north orientation gives the best results for all scenarios, regardless of the U-shape depth; we notice a reduction in the total heating and cooling consumption which vary from 0.3 to 14% (Table 5). Furthermore, the radiative and convective heat transfer coefficient exchanges increases in the U-shape wall resulting from the added adjacent surfaces have a significant effect on energy consumption in the studied building. These variations are due to the increased shadow area as well as increased natural ventilation which is more beneficial to reduce building energy consumption (see Figs. 8, 9, 10, and 11).
According to the results of Table 8, it can be clear that the optimal value for the U-shape depth (corresponding to the maximum reduction of the total heating and cooling load) is 0.5 m for the south-east and the south-west orientations and 0.75m for the north orientation in all climate zones. But in the south orientation, we observe that we have two optimal values of the U-shape depth which are 0.25 m (for Tangier, Ifrane, and Errachidia) and 0.5 m (for Meknes, Marrakech, and Rabat). Table 8 resumes these values for the six climate zones.
As preceded, it can be observed that the integration of the U-shape, regardless of the orientation of the facade concerned, introduces a decrease in the total heating and cooling load. In this part of the results, a numerical model of the external wall was developed using Comsol Multiphysics (Figs. 8 and 9) to explain the effect of different U-shape (0.25, 0.5, 0.75, 1, and 1.25) on reducing the heating and cooling loads. In the thermal point of view, this is due to a new phenomenon appearing with the U-shape as follows: • The additional radiative exchanges in the U-shape wall resulting from the added adjacent surfaces as can be concluded from the results of Fig. 11. The radiative heat transfer coefficient increases by increasing the U-shape until the value of 0.75 m and decreases slowly after that. • The modification of the convection exchanges between the wall and the exterior results from the low mobility of the air in the external part of the U-shape (Fig. 11). • The temperature variation of the external wall increases by increasing the depth from 0.25 to 1 m (Fig. 10).
These new exchanges will change the energy balance in favor of certain U-shape depths as mentioned in Tables 5, 6, and 7; the results of simulations show that the thermal exchanges induced by the U-shape give a reduction in consumption for the majority of scenarios (90 scenarios) and increases the consumption in others (30 scenarios).

Conclusion
In this paper, we have discussed a parametric study of a building model located in Morocco for a year in order to evaluate the effect of the U-shape exterior walls on the total heating and cooling load. In fact, we proposed five scenarios (different depths of the U-shape) for four orientations (south, south-east, south-west, and north). To summarize, the proposed developed study rests on the following points: • The present results are valid particularly in the Mediterranean periphery of North Africa. • The U-shape affects directly the total heating and cooling load of the building in all orientations. • A reduction in the total heating and cooling consumption is noticeable in all climate zones and vary from 0.3 to 14%.  • The U-shape is more profitable in cold to moderate climates. • The increase in convective and radiative heat transfer coefficients at different U-shape depths reduces the total heating and cooling consumption.
Finally, the U-shape exterior walls affect the heating and cooling loads of the building model depending on the orientation. The optimal orientations concerning this parameter are the north and south. These results are