Persian
Language
Geographical Research is Published in Persian Fulltext.
Volume 39, Issue 3 (2024)
GeoRes 2024, 39(3): 299-308 |
Back to browse issues page
Article Type:
Subject:
Research code: ART-1634
History
Received: 2024/07/27 | Accepted: 2024/08/28 | Published: 2024/09/28
Received: 2024/07/27 | Accepted: 2024/08/28 | Published: 2024/09/28
How to cite this article
Heydaripour O, Taher Tolou Del M. Energy Performance Modeling of Residential Isfahan Apartments Based on Geometric Characteristics. GeoRes 2024; 39 (3) :299-308
URL: http://georesearch.ir/article-1-1634-en.html
URL: http://georesearch.ir/article-1-1634-en.html
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Rights and permissions
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Authors
O. Heydaripour1, M.S. Taher Tolou Del *2
1- Department of Architecture, Faculty of Architectural Engineering and Urban Design, Shahid Rajaee Teacher Training University, Tehran, Iran
2- Department of Architecture, Faculty of Architectural Engineering and Urban Design, Shahid Rajaee Teacher Training University, Lavizan, Tehran, Iran
2- Department of Architecture, Faculty of Architectural Engineering and Urban Design, Shahid Rajaee Teacher Training University, Lavizan, Tehran, Iran
Full-Text (HTML) (7 Views)
Background
The building sector is one of the largest energy consumers in the world, and population growth has intensified the energy consumption crisis. In the city of Isfahan, residential energy consumption has also increased significantly, and non-standard designs are common. Therefore, modeling building energy performance based on physical characteristics has become a necessity for urban planning.
Previous Studies
Previous studies have shown a direct relationship between the physical factors of buildings and their energy performance (Wei et al., 2018; Liu et al., 2023). In recent decades, researchers have made significant efforts to develop energy consumption prediction models (Jin et al., 2023; Liu et al., 2023), but most of these models have not been user-friendly for non-experts. Building energy prediction approaches are generally divided into three main categories: white-box models based on physical equations (Ali et al., 2024), black-box models relying on historical data (Mousavi et al., 2023; Pan et al., 2023), and gray-box hybrid models that aim to combine the advantages of both methods (Asad Poor et al., 2021; Liu et al., 2019). Although various tools have been used for energy simulation, there is still no consensus on the most suitable method for different conditions (Olu-Ajayi et al., 2023; Gao et al., 2019; Kabir et al., 2020).
Aim(s)
The aim of this study is to model and examine the complex relationship between the geometric characteristics of buildings and their energy performance, in order to provide a practical foundation for architects, urban planners, and policymakers.
Research Type
This study is of an applied nature.
Research Society, Place and Time
The statistical population of this study consisted of physical simulation models of five-story residential buildings with pilots in the city of Isfahan, with floor areas ranging from 110 to 210 square meters. This building type was selected as a common model of residential apartments in Isfahan based on data from the Iranian Statistics Center and the Engineering Organization. The study was conducted in the year 2024-2025..
Sampling Method and Number
In the regression modeling phase, purposeful sampling was employed. A total of 2,688 physical building models were generated by systematically varying geometric parameters such as side dimensions, unit heights, orientation, and floor area. In the optimization phase, random sampling based on an evolutionary algorithm was used. For each of the three floor area ranges (110, 160, and 210 square meters), 3,000 physical models were created across 100 generations with a population size of 30 per generation, totaling 9,000 physical models evaluated for optimization. Additionally, to identify residential patterns, 390 residents of Isfahan were surveyed using random sampling.
Used Devices & Materials
In this study, energy performance modeling and simulation were conducted using Rhino version 8 and the Ladybug Tools plugin version 1.8. Optimization was performed with the help of the Galapagos and Wallacei plugins, and statistical analysis was carried out using SPSS version 27. The materials used in the models included a joist and polystyrene block structural system, exterior walls made of refractory and clay bricks, cement and gypsum mortar, and roofs and floors composed of a combination of foam concrete, reinforced concrete, ceramic, and travertine stone. These materials were incorporated as fixed parameters in the algorithm based on the most common construction methods in Isfahan.
Findings by Text
The present study aimed to evaluate the energy simulation performance in residential buildings. In the first phase, the simulation model was validated by comparing its results with actual energy consumption data from 16 residential units in the city of Isfahan. The results showed that the average difference between the actual and simulated total energy consumption was -841 kWh, and the average difference per square meter was -1.475 kWh. Additionally, the average percentage difference between simulated and actual energy consumption was approximately 3.3% (Figure 1). The correlation test between the simulated and actual values also indicated a significance level below 0.05 and a correlation coefficient above 0.75, demonstrating the good validity of the simulation model.
Figure 1. Comparative analysis of actual and simulated energy consumption per square meter of selected residential units
Next, by conducting 2,688 simulations based on parameter stepping, a correlation test was performed between the building's geometric parameters and energy performance indicators. The findings showed the highest correlation between building volume and total energy consumption with a coefficient of 0.991, followed by the floor area with a coefficient of 0.955. Thus, the sensitivity of the algorithm outputs to the inputs was confirmed. To examine the relative role of geometric parameters, linear regression analysis was employed (Table 1).
Table 1. Results of correlation test between geometric factors and simulation performance objectives
In the regression analysis of energy consumption per square meter, an R² value of 0.949 was obtained. Three variables—floor height, south window width, and south window height—were removed from the model due to high correlation with other variables. The results showed that the greatest impact was attributed to building volume and floor area, respectively, while orientation and north window width had the least effect (Table 2). The final equation for energy consumption per square meter was derived as follows:
GEC/m²=199.273+(-3.285×EWD)+(-3.150×NSD)+(0.011×BD)+(0.183×BA)+(-0.013×BV)+(0.501×WNW)+(52.350×HNW)
Table 2. Results of linear regression analysis between energy consumption per square Meter and geometric factors
In the regression model related to cooling demand per square meter, the R² value was calculated as 0.879. Similar to the previous analysis, three parameters including floor height and southern window dimensions were excluded from the model. The greatest contribution in this model belonged to the building volume and footprint area, while the least effect was attributed to orientation and the east-west dimension (Table 3). The resulting regression equation was as follows:
CED/m²=89.303+(1.122×EWD)+(-2.745×NSD)+(0.008×BD)+(0.127×BA)+(-0.009×BV)+(-2.573×WNW)+(21.464×HNW)
Table 3. Results of linear regression analysis for cooling demand per square meter and geometric parameters
The regression analysis for heating demand also yielded an R² value of 0.923. Three parameters including floor height, as well as the width and height of the south-facing window were again excluded from the model. In this model, the greatest impact was attributed to the east-west dimensions and the height of the north-facing window, while the least influence came from orientation and north-south dimensions (Table 4). The final equation was derived as follows:
HED/m²=35.766+(-4.407×EWD)+(-0.407×NSD)+(0.002×BD)+(0.056×BA)+(-0.005×BV)+(3.074×WNW)+(30.886×HNW)
Table 4. Results of linear regression analysis for heating demand per square meter and geometric factors
The results obtained from these three equations demonstrated that, on average, the models were capable of predicting over 90% of the energy performance of five-story south-facing residential buildings in the city of Isfahan which is an impressive level of success.
In the next stage, three optimization processes were carried out using the NSGA-II genetic algorithm for three building base areas: 110, 160, and 210 square meters. In each process, 30 Pareto front solutions from the 99th generation were selected, illustrating the trade-off between cooling and heating demands. For each area, the solutions were categorized into three performance classes (Figure 2).
Figure 2) Classified characteristics of pareto front responses in the three conducted optimization tests
A comparison of the average optimization results revealed that the building’s floor area, length-to-width ratio, and volume were among the most influential factors affecting energy performance. Increasing the building's east-west elongation led to a reduction in heating demand, while north-south elongation improved daylight performance and reduced cooling demand. Buildings with a square plan and minimal elongation in the north-south direction demonstrated better overall energy performance. Additionally, increasing east-west elongation in urban blocks contributed to lower energy consumption per square meter. Overall, blocks with east-west orientation performed better than those oriented north-south.
Moreover, the findings indicated that increasing vertical density and reducing the interior height to approximately 2.6 meters significantly contributed to energy optimization. For all analyzed floor area sizes, the best-performing models featured this specific interior height. Additionally, a zero-degree orientation angle was consistently identified as the most energy-efficient in terms of total energy consumption, energy use per square meter, and useful daylight illuminance. Orientations with negative angles relative to zero also outperformed their positive-angle counterparts.
The results highlight that architecture based on appropriate elongation, a zero-degree orientation, and optimal density can play a key role in the sustainable design of residential buildings (Figure 3).
Figure 3. Optimal responses of the three optimization processes based on the three energy performance objectives
Main Comparisons to Similar Studies
In this study, the modeling of the relationship between the geometric characteristics of residential apartments and energy performance was compared with the results of similar research. The current findings showed good consistency with those of Dino (2016) in terms of modeling equations, although some differences were observed regarding the use of dimensions based on orientation and building form. While the study by Souza and Alsaadani (2012) identify east-west elongation as the most influential factor, this research highlighted the impact of maximum floor area, likely due to differences in the studied climate. Unlike the study by Dino and Ocoluk (2017), which consider the effect of interior spaces, this factor was not included in the present research. The examined geometric variations were able to reduce energy consumption by up to 50%, aligning with the results of Musau and Steemers (2008), but differing from Poirazis et al. (2008), who believe that reductions beyond 40% were unlikely. Moreover, the improvements in cooling and heating demands were reported to be higher than in some previous studies. The total number of simulations conducted in this research (over 11,000) was also significantly greater and more comprehensive compared to similar studies, such as those by Yi (2016) and Dino & Ocoluk (2017).
Suggestions
It is recommended that future modeling and optimization efforts be based on variations in the number of floors and the building’s floor area. In this regard, the investigation of the minimum and maximum threshold limits for changes in both the number of floors and the building's floor area should be taken into consideration.
Conclusion
The geometric characteristics of apartment housing have a significant impact on building energy consumption and, consequently, on global energy use. Among the geometric factors, the building’s floor area, length-to-width ratio, and volume have the greatest influence on the energy performance of five-story residential buildings, leading to reduced energy consumption. The height of interior spaces is also an important factor affecting energy use, where its reduction results in lower energy consumption.
Acknowledgments: None reported by the authors.
Ethical Permission: None reported by the authors.
Conflict of Interest: None reported by the authors.
Authors’ Contributions: Heydaripour O (First author), Main Researcher/Introduction Writer/Statistical Analyst (50%); Taher Tolou Del MS (Second author), Discussion Writer/Methodologist (50%).
Funding: None reported by the authors.
The building sector is one of the largest energy consumers in the world, and population growth has intensified the energy consumption crisis. In the city of Isfahan, residential energy consumption has also increased significantly, and non-standard designs are common. Therefore, modeling building energy performance based on physical characteristics has become a necessity for urban planning.
Previous Studies
Previous studies have shown a direct relationship between the physical factors of buildings and their energy performance (Wei et al., 2018; Liu et al., 2023). In recent decades, researchers have made significant efforts to develop energy consumption prediction models (Jin et al., 2023; Liu et al., 2023), but most of these models have not been user-friendly for non-experts. Building energy prediction approaches are generally divided into three main categories: white-box models based on physical equations (Ali et al., 2024), black-box models relying on historical data (Mousavi et al., 2023; Pan et al., 2023), and gray-box hybrid models that aim to combine the advantages of both methods (Asad Poor et al., 2021; Liu et al., 2019). Although various tools have been used for energy simulation, there is still no consensus on the most suitable method for different conditions (Olu-Ajayi et al., 2023; Gao et al., 2019; Kabir et al., 2020).
Aim(s)
The aim of this study is to model and examine the complex relationship between the geometric characteristics of buildings and their energy performance, in order to provide a practical foundation for architects, urban planners, and policymakers.
Research Type
This study is of an applied nature.
Research Society, Place and Time
The statistical population of this study consisted of physical simulation models of five-story residential buildings with pilots in the city of Isfahan, with floor areas ranging from 110 to 210 square meters. This building type was selected as a common model of residential apartments in Isfahan based on data from the Iranian Statistics Center and the Engineering Organization. The study was conducted in the year 2024-2025..
Sampling Method and Number
In the regression modeling phase, purposeful sampling was employed. A total of 2,688 physical building models were generated by systematically varying geometric parameters such as side dimensions, unit heights, orientation, and floor area. In the optimization phase, random sampling based on an evolutionary algorithm was used. For each of the three floor area ranges (110, 160, and 210 square meters), 3,000 physical models were created across 100 generations with a population size of 30 per generation, totaling 9,000 physical models evaluated for optimization. Additionally, to identify residential patterns, 390 residents of Isfahan were surveyed using random sampling.
Used Devices & Materials
In this study, energy performance modeling and simulation were conducted using Rhino version 8 and the Ladybug Tools plugin version 1.8. Optimization was performed with the help of the Galapagos and Wallacei plugins, and statistical analysis was carried out using SPSS version 27. The materials used in the models included a joist and polystyrene block structural system, exterior walls made of refractory and clay bricks, cement and gypsum mortar, and roofs and floors composed of a combination of foam concrete, reinforced concrete, ceramic, and travertine stone. These materials were incorporated as fixed parameters in the algorithm based on the most common construction methods in Isfahan.
Findings by Text
The present study aimed to evaluate the energy simulation performance in residential buildings. In the first phase, the simulation model was validated by comparing its results with actual energy consumption data from 16 residential units in the city of Isfahan. The results showed that the average difference between the actual and simulated total energy consumption was -841 kWh, and the average difference per square meter was -1.475 kWh. Additionally, the average percentage difference between simulated and actual energy consumption was approximately 3.3% (Figure 1). The correlation test between the simulated and actual values also indicated a significance level below 0.05 and a correlation coefficient above 0.75, demonstrating the good validity of the simulation model.
Figure 1. Comparative analysis of actual and simulated energy consumption per square meter of selected residential units
Next, by conducting 2,688 simulations based on parameter stepping, a correlation test was performed between the building's geometric parameters and energy performance indicators. The findings showed the highest correlation between building volume and total energy consumption with a coefficient of 0.991, followed by the floor area with a coefficient of 0.955. Thus, the sensitivity of the algorithm outputs to the inputs was confirmed. To examine the relative role of geometric parameters, linear regression analysis was employed (Table 1).
Table 1. Results of correlation test between geometric factors and simulation performance objectives
In the regression analysis of energy consumption per square meter, an R² value of 0.949 was obtained. Three variables—floor height, south window width, and south window height—were removed from the model due to high correlation with other variables. The results showed that the greatest impact was attributed to building volume and floor area, respectively, while orientation and north window width had the least effect (Table 2). The final equation for energy consumption per square meter was derived as follows:
GEC/m²=199.273+(-3.285×EWD)+(-3.150×NSD)+(0.011×BD)+(0.183×BA)+(-0.013×BV)+(0.501×WNW)+(52.350×HNW)
Table 2. Results of linear regression analysis between energy consumption per square Meter and geometric factors
In the regression model related to cooling demand per square meter, the R² value was calculated as 0.879. Similar to the previous analysis, three parameters including floor height and southern window dimensions were excluded from the model. The greatest contribution in this model belonged to the building volume and footprint area, while the least effect was attributed to orientation and the east-west dimension (Table 3). The resulting regression equation was as follows:
CED/m²=89.303+(1.122×EWD)+(-2.745×NSD)+(0.008×BD)+(0.127×BA)+(-0.009×BV)+(-2.573×WNW)+(21.464×HNW)
Table 3. Results of linear regression analysis for cooling demand per square meter and geometric parameters
The regression analysis for heating demand also yielded an R² value of 0.923. Three parameters including floor height, as well as the width and height of the south-facing window were again excluded from the model. In this model, the greatest impact was attributed to the east-west dimensions and the height of the north-facing window, while the least influence came from orientation and north-south dimensions (Table 4). The final equation was derived as follows:
HED/m²=35.766+(-4.407×EWD)+(-0.407×NSD)+(0.002×BD)+(0.056×BA)+(-0.005×BV)+(3.074×WNW)+(30.886×HNW)
Table 4. Results of linear regression analysis for heating demand per square meter and geometric factors
The results obtained from these three equations demonstrated that, on average, the models were capable of predicting over 90% of the energy performance of five-story south-facing residential buildings in the city of Isfahan which is an impressive level of success.
In the next stage, three optimization processes were carried out using the NSGA-II genetic algorithm for three building base areas: 110, 160, and 210 square meters. In each process, 30 Pareto front solutions from the 99th generation were selected, illustrating the trade-off between cooling and heating demands. For each area, the solutions were categorized into three performance classes (Figure 2).
Figure 2) Classified characteristics of pareto front responses in the three conducted optimization tests
A comparison of the average optimization results revealed that the building’s floor area, length-to-width ratio, and volume were among the most influential factors affecting energy performance. Increasing the building's east-west elongation led to a reduction in heating demand, while north-south elongation improved daylight performance and reduced cooling demand. Buildings with a square plan and minimal elongation in the north-south direction demonstrated better overall energy performance. Additionally, increasing east-west elongation in urban blocks contributed to lower energy consumption per square meter. Overall, blocks with east-west orientation performed better than those oriented north-south.
Moreover, the findings indicated that increasing vertical density and reducing the interior height to approximately 2.6 meters significantly contributed to energy optimization. For all analyzed floor area sizes, the best-performing models featured this specific interior height. Additionally, a zero-degree orientation angle was consistently identified as the most energy-efficient in terms of total energy consumption, energy use per square meter, and useful daylight illuminance. Orientations with negative angles relative to zero also outperformed their positive-angle counterparts.
The results highlight that architecture based on appropriate elongation, a zero-degree orientation, and optimal density can play a key role in the sustainable design of residential buildings (Figure 3).
Figure 3. Optimal responses of the three optimization processes based on the three energy performance objectives
Main Comparisons to Similar Studies
In this study, the modeling of the relationship between the geometric characteristics of residential apartments and energy performance was compared with the results of similar research. The current findings showed good consistency with those of Dino (2016) in terms of modeling equations, although some differences were observed regarding the use of dimensions based on orientation and building form. While the study by Souza and Alsaadani (2012) identify east-west elongation as the most influential factor, this research highlighted the impact of maximum floor area, likely due to differences in the studied climate. Unlike the study by Dino and Ocoluk (2017), which consider the effect of interior spaces, this factor was not included in the present research. The examined geometric variations were able to reduce energy consumption by up to 50%, aligning with the results of Musau and Steemers (2008), but differing from Poirazis et al. (2008), who believe that reductions beyond 40% were unlikely. Moreover, the improvements in cooling and heating demands were reported to be higher than in some previous studies. The total number of simulations conducted in this research (over 11,000) was also significantly greater and more comprehensive compared to similar studies, such as those by Yi (2016) and Dino & Ocoluk (2017).
Suggestions
It is recommended that future modeling and optimization efforts be based on variations in the number of floors and the building’s floor area. In this regard, the investigation of the minimum and maximum threshold limits for changes in both the number of floors and the building's floor area should be taken into consideration.
Conclusion
The geometric characteristics of apartment housing have a significant impact on building energy consumption and, consequently, on global energy use. Among the geometric factors, the building’s floor area, length-to-width ratio, and volume have the greatest influence on the energy performance of five-story residential buildings, leading to reduced energy consumption. The height of interior spaces is also an important factor affecting energy use, where its reduction results in lower energy consumption.
Acknowledgments: None reported by the authors.
Ethical Permission: None reported by the authors.
Conflict of Interest: None reported by the authors.
Authors’ Contributions: Heydaripour O (First author), Main Researcher/Introduction Writer/Statistical Analyst (50%); Taher Tolou Del MS (Second author), Discussion Writer/Methodologist (50%).
Funding: None reported by the authors.
References
1. Al-Shargabi AA, Almhafdy A, Ibrahim DM, Alghieth M, Chiclana F (2022). Buildings' energy consumption prediction models based on buildings' characteristics: Research trends, taxonomy, and performance measures. Journal of Building Engineering. 54:104577. [Link] [DOI:10.1016/j.jobe.2022.104577]
2. Ali A, Jayaraman R, Azar E, Maalouf M (2024). A comparative analysis of machine learning and statistical methods for evaluating building performance: A systematic review and future benchmarking framework. Building and Environment. 252:111268. [Link] [DOI:10.1016/j.buildenv.2024.111268]
3. Asad Poor J, Goh Y W, Thorpe D (2021). A human-centric participatory approach to energy-efficient housing based on occupants' collaborative image. Open House International. 46(4):615-635. [Link] [DOI:10.1108/OHI-11-2020-0163]
4. Chen Y, Guo M, Chen Z, Chen Z, Ji Y (2022). Physical energy and data-driven models in building energy prediction: A review. Energy Reports. 8:2656-2671. [Link] [DOI:10.1016/j.egyr.2022.01.162]
5. Chong A, Augenbroe G, Yan D (2021). Occupancy data at different spatial resolutions: Building energy performance and model calibration. Applied Energy. 286:116492. [Link] [DOI:10.1016/j.apenergy.2021.116492]
6. Coakley D, Raftery P, Keane M (2014). A review of methods to match building energy simulation models to measured data. Renewable and Sustainable Energy Reviews. 37:123-141. [Link] [DOI:10.1016/j.rser.2014.05.007]
7. Dino I. (2016). An evolutionary approach for 3D architectural space layout design exploration. Automation in construction. 69:131-150. [Link] [DOI:10.1016/j.autcon.2016.05.020]
8. Dino I G, Üçoluk G (2017). Multiobjective design optimization of building space layout, energy, and daylighting performance. Journal of Computing in Civil Engineering. 31(5):04017025. [Link] [DOI:10.1061/(ASCE)CP.1943-5487.0000669]
9. Ekici B, Cubukcuoglu C, Turrin M, Sariyildiz IS (2019). Performative computational architecture using swarm and evolutionary optimisation: A review. Building and Environment. 147:356-371. [Link] [DOI:10.1016/j.buildenv.2018.10.023]
10. Elbeltagi E, Wefki H (2021). Predicting energy consumption for residential buildings using ANN through parametric modeling. Energy Reports. 7:2534-2545. [Link] [DOI:10.1016/j.egyr.2021.04.053]
11. Feng Z, Zhang M, Wei N, Zhao J, Zhang T, He X (2022). An office building energy consumption forecasting model with dynamically combined residual error correction based on the optimal model. Energy Reports. 8:12442-12455. [Link] [DOI:10.1016/j.egyr.2022.09.022]
12. Gao H, Koch C, Wu Y (2019). Building information modelling based building energy modelling: A review. Applied energy. 238:320-343. [Link] [DOI:10.1016/j.apenergy.2019.01.032]
13. González-Torres M, Pérez-Lombard L, Coronel J F, Maestre I R, Yan D. (2022). A review on buildings energy information: Trends, end-uses, fuels and drivers. Energy Reports. 8:626-637. [Link] [DOI:10.1016/j.egyr.2021.11.280]
14. Jin X, Zhang C, Xiao F, Li A, Miller C (2023). A review and reflection on open datasets of city-level building energy use and their applications. Energy and Buildings. 285:112911. [Link] [DOI:10.1016/j.enbuild.2023.112911]
15. Kabir S, Islam RU, Hossain MS, Andersson K (2020). An integrated approach of belief rule base and deep learning to predict air pollution. Sensors. 20(7):1956. [Link] [DOI:10.3390/s20071956]
16. Liu H, Liang J, Liu Y, Wu H (2023). A review of data-driven building energy prediction. Buildings. 13(2):532. [Link] [DOI:10.3390/buildings13020532]
17. Liu K, Xu X, Zhang R, Kong L, Wang W, Deng W (2023). Impact of urban form on building energy consumption and solar energy potential: A case study of residential blocks in Jianhu, China. Energy and Buildings: 280:112727. [Link] [DOI:10.1016/j.enbuild.2022.112727]
18. Liu XH, Zhang DG, Yan HR, Cui YY, Chen L (2019). A new algorithm of the best path selection based on machine learning. IEEE Access. 7:126913-126928. [Link] [DOI:10.1109/ACCESS.2019.2939423]
19. Mousavi S, Villarreal-Marroquín MG, Hajiaghaei-Keshteli M, Smith NR (2023). Data-driven prediction and optimization toward net-zero and positive-energy buildings: A systematic review. Building and Environment. 242:110578. [Link] [DOI:10.1016/j.buildenv.2023.110578]
20. Musau F, Steemers K (2008). Space planning and energy efficiency in office buildings: the role of spatial and temporal diversity. Architectural Science Review. 51(2):133-145. [Link] [DOI:10.3763/asre.2008.5117]
21. Olu-Ajayi R, Alaka H, Sulaimon I, Balogun H, Wusu G, Yusuf W, et al (2023). Building energy performance prediction: A reliability analysis and evaluation of feature selection methods. Expert Systems with Applications. 225:120109. [Link] [DOI:10.1016/j.eswa.2023.120109]
22. Olu-Ajayi R, Alaka H, Sulaimon I, Sunmola F, Ajayi S (2022). Machine learning for energy performance prediction at the design stage of buildings. Energy for Sustainable Development. 66:12-25. [Link] [DOI:10.1016/j.esd.2021.11.002]
23. Pan Y, Zhu M, Lv Y, Yang Y, Liang Y, Yin R, et al (2023). Building energy simulation and its application for building performance optimization: A review of methods, tools, and case studies. Advances in Applied Energy. 10:100135. [Link] [DOI:10.1016/j.adapen.2023.100135]
24. Poirazis H (2008). Single and double skin glazed office buildings. Division of Energy and Building Design [Report]. Lund: Lund University. [Link]
25. Sariyildiz S (2012). Performative computational design. Proceedings of the ICONARCH-International Congress of Architecture and Technology; 2012 Nov 15-17. Konya: Selcuk University. [Link]
26. Souza C, Alsaadani S (2012). Thermal zoning in speculative office buildings: discussing the connections between space layout and inside temperature control. Proceedings of the 1st Building Simulation and Optimization Conference; 2012; Loughborough University, England. Loughborough: IBPSA Publication. [Link]
27. Wei S, Jones R, De Wilde P (2014). Driving factors for occupant-controlled space heating in residential buildings. Energy and Buildings. 70:36-44. [Link] [DOI:10.1016/j.enbuild.2013.11.001]
28. Wei Y, Zhang X, Shi Y, Xia L, Pan S, Wu J, et al (2018). A review of data-driven approaches for prediction and classification of building energy consumption. Renewable and Sustainable Energy Reviews. 82(1):1027-1047. [Link] [DOI:10.1016/j.rser.2017.09.108]
29. Yi H (2016). User-driven automation for optimal thermal-zone layout during space programming phases. Architectural Science Review. 59(4):279-306. [Link] [DOI:10.1080/00038628.2015.1021747]
30. Zhang Y, Bai X, Mills FP, Pezzey JC (2018). Rethinking the role of occupant behavior in building energy performance: A review. Energy and Buildings. 172:279-294. [Link] [DOI:10.1016/j.enbuild.2018.05.017]
31. Zhuang D, Zhang X, Lu Y, Wang C, Jin X, Zhou X, Shi X (2021). A performance data integrated BIM framework for building life-cycle energy efficiency and environmental optimization design. Automation in Construction. 127:103712. [Link] [DOI:10.1016/j.autcon.2021.103712]