Journal of the Architectural Institute of Korea (대한건축학회논문집)

Architectural Institute of Korea (대한건축학회)
권호 표지
DOI QR코드

DOI QR Code

Review of Machine Learning for Building Energy Prediction

건물에너지 예측을 위한 기계학습 모델 검토

  • Kwon, Oh Ik (R&D Institute, Hanil Mechanical Electrical Consultants) ;
  • Kim, Young Il (Dept. of Architectural Engineering, Seoul National University of Science and Technology)
  • 권오익 ((주)한일엠이씨 연구개발본부) ;
  • 김영일 (서울과학기술대학교)
  • Received : 2022.12.14
  • Accepted : 2023.04.10
  • Published : 2023.05.30
⟨ Previous Next ⟩

Abstract

To prepare basic data for the use of machine learning in the building energy field, this study examined the characteristics of each model and compared the prediction performance, calculation efficiency and output result aspects of the machine learning model according to the input parameters. Outdoor temperature was used as a basic input to consider input differences for six machine learning models, MLR, SVM, GPR, ANN, DNN and DT, which are mainly used in the building energy field, and the building energy consumption was predicted and compared depending on whether the indoor temperature was additionally reflected. The predictive performance of most models improved when the outdoor temperature and the indoor temperature were reflected as inputs rather than when the outdoor temperature was reflected as an input in the influence of the input parameters. In the comparison of the predictive performance of the model, DNN(5-Layer) showed the most dominant predictive results with RMSE, MSE, MAE, and R2 (0.190, 0.036, 0.139, 0.88). Next, ANN showed predictive performance of RMSE, MSE, MAE, R2 (0.203, 0.041, 0.142, 0.86), and GPR provided efficient prediction with RMSE, MSE, MAE, R2 (0.211, 0.044, 0.150, 0.85). DNN and ANN improved their prediction performance as the number of hidden layers increased, but the training time increased from 4.8 seconds to 16.5 seconds. In terms of computational efficiency considering training time, MLR showed the best result with 1.4s. As a result, DNN showed 14% better predictive performance than MLR, and MLR were trained 11.8 times faster than DNN. With indoor temperature being further reflected as input parameters, most models better represent actual building energy consumption in aspects of the forecast results. Machine learning model selection should be reviewed not only for predictive performance for errors but also for calculation cost and the discernment provided by predictive results. Since this study was conducted on a single building, research on the selection and development of models with high reproducibility in various models based on big data in terms of utilization should be continued.

Keywords

Acknowledgement

본 연구는 2022년도 국토교통부 도시건축연구개발사업 연구비지원에 의한 결과의 일부임. 과제번호: 22AUDP-C151656-04

References

  1. Ahmad, A. S., Hassan, M. Y., Abdullah, M. P., Rahman, H. A., Hussin, F., Abdullah, H., & Saidur, R. (2014). A review on applications of ANN and SVM for building electrical energy consumption forecasting. Renewable and Sustainable Energy Reviews, 33, 102-109. https://doi.org/10.1016/j.rser.2014.01.069
  2. Amasyali, K., & El-Gohary, N. M. (2018). A review of data-driven building energy consumption prediction studies. Renewable and Sustainable Energy Reviews, 81, 1192-1205. https://doi.org/10.1016/j.rser.2017.04.095
  3. Beckman, W. A., Broman, L., Fiksel, A., Klein, S. A., Lindberg, E., Schuler, M., & Thornton, J. (1994). TRNSYS The most complete solar energy system modeling and simulation software, Renewable energy, 5(1-4), 486-488. https://doi.org/10.1016/0960-1481(94)90420-0
  4. Bourhnane, S., Abid, M. R., Lghoul, R., Zine-Dine, K., Elkamoun, N., & Benhaddou, D. (2020). Machine learning for energy consumption prediction and scheduling in smart buildings. SN Applied Sciences, 2(2), 1-10. https://doi.org/10.1007/s42452-019-1685-8
  5. Cholewa, T., Siuta-Olcha, A., Smolarz, A., Muryjas, P., Wolszczak, P., Anasiewicz, R., & Balaras, C. A. (2021). A simple building energy model in form of an equivalent outdoor temperature. Energy and Buildings, 236, 110766.
  6. Chou, J. S., & Bui, D. K. (2014). Modeling heating and cooling loads by artificial intelligence for energy-efficient building design, Energy and Buildings, 82, 437-446. https://doi.org/10.1016/j.enbuild.2014.07.036
  7. Crawley, D. B., Lawrie, L. K., Winkelmann, F. C., Buhl, W. F., Huang, Y. J., Pedersen, C. O., ... & Glazer, J. (2001). EnergyPlus: creating a new-generation building energy simulation program, Energy and Buildings, 33(4), 319-331. https://doi.org/10.1016/S0378-7788(00)00114-6
  8. Gundersen, O. E., Shamsaliei, S., & Isdahl, R. J. (2022). Do machine learning platforms provide out-of-the-box reproducibility?, Future Generation Computer Systems, 126, 34-47. https://doi.org/10.1016/j.future.2021.06.014
  9. Lee, S., Jung, S., & Lee, J. (2019). Prediction model based on an artificial neural network for user-based building energy consumption in South Korea. Energies, 12(4), 608.
  10. Lei, L., Chen, W., Wu, B., Chen, C., & Liu, W. (2021). A building energy consumption prediction model based on rough set theory and deep learning algorithms. Energy and Buildings, 240, 110886.
  11. Li, L., Sun, W., Hu, W., & Sun, Y. (2021). Impact of natural and social environmental factors on building energy consumption: Based on bibliometrics, Journal of Building Engineering, 37, 102136.
  12. Li, Z., Han, Y., & Xu, P. (2014). Methods for benchmarking building energy consumption against its past or intended performance: An overview. Applied Energy, 124, 325-334.
  13. Lu, C., Li, S., & Lu, Z. (2021). Building energy prediction using artificial neural networks: A literature survey. Energy and Buildings, 111718.
  14. Moon, J., Park, S., Rho, S., & Hwang, E. (2019). A comparative analysis of artificial neural network architectures for building energy consumption forecasting. International Journal of Distributed Sensor Networks, 15(9), 1550147719877616.
  15. Olu-Ajayi, R., Alaka, H., Sulaimon, I., Sunmola, F., & Ajayi, S. (2022). Building energy consumption prediction for residential buildings using deep learning and other machine learning techniques, Journal of Building Engineering, 45, 103406.
  16. Seyedzadeh, S., Rahimian, F. P., Glesk, I., & Roper, M. (2018). Machine learning for estimation of building energy consumption and performance: a review. Visualization in Engineering, 6(1), 1-20. https://doi.org/10.1186/s40327-018-0062-9
  17. Shapi, M. K. M., Ramli, N. A., & Awalin, L. J. (2021). Energy consumption prediction by using machine learning for smart building: case study in Malaysia. Developments in the Built Environment, 5, 100037.
  18. Wang, S., Yan, C., & Xiao, F. (2012). Quantitative energy performance assessment methods for existing buildings, Energy and Buildings, 55, 873-888. https://doi.org/10.1016/j.enbuild.2012.08.037
  19. Wang, W., Guo, H., Li, X., Tang, S., Xia, J., & Lv, Z. (2022). Deep learning for assessment of environmental satisfaction using BIM big data in energy efficient building digital twins. Sustainable Energy Technologies and Assessments, 50, 101897.
  20. Wei, Y., Zhang, X., Shi, Y., Xia, L., Pan, S., Wu, J., ... & Zhao, X. (2018). A review of data-driven approaches for prediction and classification of building energy consumption, Renewable and Sustainable Energy Reviews, 82, 1027-1047. https://doi.org/10.1016/j.rser.2017.09.108
  21. Yoon, J., Lee, B., Yeo, M., & Kim, K. (2017). Classification of energy consumption patterns in university buildings using change point model and analysis according to energy impact factors, Journal of the Architectural institute of Korea structure & construction, 33(11), 71-78. https://doi.org/10.5659/JAIK_SC.2017.33.2.71
  22. Yu, Z., Haghighat, F., Fung, B. C., & Yoshino, H. (2010). A decision tree method for building energy demand modeling. Energy and Buildings, 42(10), 1637-1646.  https://doi.org/10.1016/j.enbuild.2010.04.006