Araştırma Makalesi
BibTex RIS Kaynak Göster

Isıtma ve Soğutma Yükü Gereksinimlerine Dayalı Bina Enerji Verimliliği Değerlendirmesi için Regresyon Modellerinin Karşılaştırmalı Değerlendirmesi

Yıl 2025, Cilt: 11 Sayı: 2, 283 - 303, 31.12.2025
https://doi.org/10.34186/klujes.1804525

Öz

Isıtma ve soğutma yüklerinin doğru tahmini, enerji verimli binalar tasarlamak ve çevresel ayak izlerini azaltmak için kritik bir ön koşuldur. Bu çalışma, mimari parametrelerine dayalı olarak konut binalarının enerji verimliliğini tahmin etmek için çoklu regresyon modellerinin kapsamlı bir karşılaştırmalı analizini sunmaktadır. Enerji Verimliliği veri setini kullanarak, yedi farklı modelleme yaklaşımının performansını değerlendirdik: Doğrusal Regresyon, Karar Ağacı, Rastgele Orman, Radyal Taban Fonksiyonu çekirdeğine sahip Destek Vektör Regresyonu, K-En Yakın Komşular, Çok Katmanlı Algılayıcı ve Derin Sinir Ağları. Modeller, Kök Ortalama Karesel Hata, Ortalama Mutlak Hata ve belirleme katsayısı (R²) kullanılarak titizlikle değerlendirildi. Sonuçlar, doğrusal olmayan makine öğrenmesi yöntemlerinin geleneksel doğrusal modellerden önemli ölçüde daha iyi performans gösterdiğini göstermektedir. Özellikle, Rastgele Orman ve Destek Vektör Regresyonu modelleri, ısıtma yükü için 0,46 ve soğutma yükü için 1,53 kadar düşük RMSE değerleri ve 0,97'yi aşan R² puanlarıyla üstün tahmin doğruluğu elde etti. Ayrıca, özellik önem analizi, ısıtma ve soğutma yükü tahminleri için sırasıyla Toplam Yükseklik ve Göreceli Kompaktlığı en etkili parametreler olarak belirleyerek mimari tasarım için uygulanabilir içgörüler sağlamıştır. Bu araştırma, özellikle Rastgele Orman ve Destek Vektör Regresyonu olmak üzere gelişmiş makine öğrenimi modellerinin, bina enerji değerlendirmesi için sağlam ve doğru bir çerçeve sunduğunu göstermiştir.

Kaynakça

  • Abdel‑Jaber, F., & Dirks, K. N. (2024). A Review of Cooling and Heating Loads Predictions of Residential Buildings Using Data‑Driven Techniques. Buildings, 14(3), 752. https://doi.org/10.3390/buildings14030752
  • Andersen, R., Fabi, V., & Corgnati, S. P. (2022). Human behavior and building performance: Understanding occupant-driven energy use. Energy and Buildings, 259, 111931.
  • Ascione, F., Bianco, N., Mauro, G. M., & Vanoli, G. P. (2023). Thermal envelope optimization for European climates. Renewable Energy, 214, 1363–1378.
  • Bertoldi, P., Mosconi, R., & Serrenho, T. (2023). Efficient heat pump systems for decarbonizing EU buildings. Energy Policy, 182, 113976.
  • Chaganti, R., Rustam, F., Daghriri, T., Díez, I. d. l. T., Mazón, J. L. V., Rodríguez, C. L., & Ashraf, I. (2022). Building Heating and Cooling Load Prediction Using Ensemble Machine Learning Model. Sensors, 22(19), 7692. https://doi.org/10.3390/s22197692
  • D’Oca, S., Hong, T., & Langevin, J. (2023). Smart building technologies for energy management in Europe. Building and Environment, 233, 110161.
  • European Commission. (2023). Renovation Wave for Europe: Greening our buildings, creating jobs, improving lives. Brussels: EC Publications.
  • European Commission. (2024). Energy Performance of Buildings Directive (Recast). Brussels: EC Publications.
  • Gou, S., Zhao, J., & Pan, W. (2022). Evaluating energy policy impacts on building retrofitting. Energy Research & Social Science, 87, 102452.
  • Guerra Santin, O., & Tweed, C. (2022). Deep renovation strategies in European housing. Energy and Buildings, 259, 111912.
  • IEA. (2024). Energy Efficiency 2024: Global Progress and European Insights. Paris: International Energy Agency.
  • Kwak, S. Y., Bertoldi, P., & Economidou, M. (2023). Barriers and drivers for energy renovation in the EU. Energy Policy, 182, 113985.
  • Lu, C., Li, S., Penaka, S., & Olofsson, T. (2023). Automated Machine Learning-Based Framework of Heating and Cooling Load Prediction for Quick Residential Building Design. Energy, 274, 127334. https://doi.org/10.1016/j.energy.2023.127334
  • Miskolc, M., et al. (2023). Prediction and Optimization of Thermal Loads in Buildings with Different Shapes by Neural Networks and Recent Finite Difference Methods. Buildings, 13(11), 2862. https://doi.org/10.3390/buildings13112862
  • Tsanas, A., & Xifara, A. (2012a). Accurate quantitative estimation of energy performance of residential buildings using statistical machine learning tools. Energy and Buildings, 49, 560–567. https://doi.org/10.1016/j.enbuild.2012.03.003
  • Tsanas, A., & Xifara, A. (2012b). Energy Efficiency [Data set]. UCI Machine Learning Repository. https://doi.org/10.24432/C51307
  • Palermo, S. A., Danza, L., & Nocera, F. (2024). Adaptive façades for Mediterranean climates. Sustainable Cities and Society, 104, 104239.
  • UCI. (2025). Energy efficiency Data Set. https://archive.ics.uci.edu/ml/datasets/energy+efficiency
  • Zhou, T., Fang, Y., & Li, Z. (2024). AI-based optimization of building energy systems. Applied Energy, 351, 121550.

Comparative Evaluation of Regression Models for Building Energy Efficiency Assessment Based on Heating and Cooling Load Requirements

Yıl 2025, Cilt: 11 Sayı: 2, 283 - 303, 31.12.2025
https://doi.org/10.34186/klujes.1804525

Öz

The accurate prediction of heating and cooling loads is a critical prerequisite for designing energy-efficient buildings and reducing their environmental footprint. This study presents a comprehensive comparative analysis of multiple regression models for estimating the energy efficiency of residential buildings based on their architectural parameters. Using the Energy Efficiency dataset, we evaluated the performance of seven distinct modelling approaches: Linear Regression, Decision Tree, Random Forest, Support Vector Regression with a Radial Basis Function kernel, K-Nearest Neighbours, Multi-Layer Perceptron, and Deep Neural Networks. Models were rigorously assessed using Root Mean Square Error, Mean Absolute Error, and the coefficient of determination (R²). The results demonstrate that non-linear machine learning methods significantly outperform traditional linear models. Specifically, the Random Forest and Support Vector Regression models achieved superior predictive accuracy, with RMSE values as low as 0.46 for heating load and 1.53 for cooling load, and R² scores exceeding 0.97. Furthermore, feature importance analysis identified Overall Height and Relative Compactness as the most influential parameters for heating and cooling load predictions, respectively, providing actionable insights for architectural design. This research shows that advanced machine learning models, particularly Random Forest and Support Vector Regression, offer a robust and accurate framework for building energy assessment.

Kaynakça

  • Abdel‑Jaber, F., & Dirks, K. N. (2024). A Review of Cooling and Heating Loads Predictions of Residential Buildings Using Data‑Driven Techniques. Buildings, 14(3), 752. https://doi.org/10.3390/buildings14030752
  • Andersen, R., Fabi, V., & Corgnati, S. P. (2022). Human behavior and building performance: Understanding occupant-driven energy use. Energy and Buildings, 259, 111931.
  • Ascione, F., Bianco, N., Mauro, G. M., & Vanoli, G. P. (2023). Thermal envelope optimization for European climates. Renewable Energy, 214, 1363–1378.
  • Bertoldi, P., Mosconi, R., & Serrenho, T. (2023). Efficient heat pump systems for decarbonizing EU buildings. Energy Policy, 182, 113976.
  • Chaganti, R., Rustam, F., Daghriri, T., Díez, I. d. l. T., Mazón, J. L. V., Rodríguez, C. L., & Ashraf, I. (2022). Building Heating and Cooling Load Prediction Using Ensemble Machine Learning Model. Sensors, 22(19), 7692. https://doi.org/10.3390/s22197692
  • D’Oca, S., Hong, T., & Langevin, J. (2023). Smart building technologies for energy management in Europe. Building and Environment, 233, 110161.
  • European Commission. (2023). Renovation Wave for Europe: Greening our buildings, creating jobs, improving lives. Brussels: EC Publications.
  • European Commission. (2024). Energy Performance of Buildings Directive (Recast). Brussels: EC Publications.
  • Gou, S., Zhao, J., & Pan, W. (2022). Evaluating energy policy impacts on building retrofitting. Energy Research & Social Science, 87, 102452.
  • Guerra Santin, O., & Tweed, C. (2022). Deep renovation strategies in European housing. Energy and Buildings, 259, 111912.
  • IEA. (2024). Energy Efficiency 2024: Global Progress and European Insights. Paris: International Energy Agency.
  • Kwak, S. Y., Bertoldi, P., & Economidou, M. (2023). Barriers and drivers for energy renovation in the EU. Energy Policy, 182, 113985.
  • Lu, C., Li, S., Penaka, S., & Olofsson, T. (2023). Automated Machine Learning-Based Framework of Heating and Cooling Load Prediction for Quick Residential Building Design. Energy, 274, 127334. https://doi.org/10.1016/j.energy.2023.127334
  • Miskolc, M., et al. (2023). Prediction and Optimization of Thermal Loads in Buildings with Different Shapes by Neural Networks and Recent Finite Difference Methods. Buildings, 13(11), 2862. https://doi.org/10.3390/buildings13112862
  • Tsanas, A., & Xifara, A. (2012a). Accurate quantitative estimation of energy performance of residential buildings using statistical machine learning tools. Energy and Buildings, 49, 560–567. https://doi.org/10.1016/j.enbuild.2012.03.003
  • Tsanas, A., & Xifara, A. (2012b). Energy Efficiency [Data set]. UCI Machine Learning Repository. https://doi.org/10.24432/C51307
  • Palermo, S. A., Danza, L., & Nocera, F. (2024). Adaptive façades for Mediterranean climates. Sustainable Cities and Society, 104, 104239.
  • UCI. (2025). Energy efficiency Data Set. https://archive.ics.uci.edu/ml/datasets/energy+efficiency
  • Zhou, T., Fang, Y., & Li, Z. (2024). AI-based optimization of building energy systems. Applied Energy, 351, 121550.
Toplam 19 adet kaynakça vardır.

Ayrıntılar

Birincil Dil İngilizce
Konular Pekiştirmeli Öğrenme
Bölüm Araştırma Makalesi
Yazarlar

Sinan Atıcı 0000-0002-5997-2969

Gürkan Tuna 0000-0002-6466-4696

Gönderilme Tarihi 15 Ekim 2025
Kabul Tarihi 5 Aralık 2025
Yayımlanma Tarihi 31 Aralık 2025
Yayımlandığı Sayı Yıl 2025 Cilt: 11 Sayı: 2

Kaynak Göster

APA Atıcı, S., & Tuna, G. (2025). Comparative Evaluation of Regression Models for Building Energy Efficiency Assessment Based on Heating and Cooling Load Requirements. Kirklareli University Journal of Engineering and Science, 11(2), 283-303. https://doi.org/10.34186/klujes.1804525