The continuous increase in energy consumption has drawn global attention to its significant environmental impact, which is exacerbated by increased greenhouse gas emissions, global warming, and rapid climate change. As a result, more energy-efficient buildings are required to reduce heating and cooling energy consumption. The current study introduces a set of machine learning-based models for predicting building heating and cooling loads. Backpropagation artificial neural network, generalized regression neural network, and so on are examples of this. Radial basis neural networks, radial kernel support vector machines, and ANOVA kernel support vector machines are all examples of support vector machines. Comparisons were made based on mean absolute percentage error (MAPE), mean absolute error (MAE), relative absolute error (RAE), root relative squared error (RRSE), and root-mean squared error (RMSE) (RMSE). The significance of the machine learning models' capacities is assessed using two-tailed student t-tests. Finally, the average ranking algorithm is used to conduct a comprehensive evaluation of the machine learning models. The results show that the radial basis function network significantly outperformed the previously mentioned machine learning models.
Author (S) DetailsEslam Mohammed Abdelkader
Structural Engineering Department, Faculty of Engineering, Cairo University, Giza, Egypt.
Abobakr Al-Sakkaf
Department of Building, Civil, and Environmental Engineering, Concordia University, Montreal, QC, Canada and Department of Architecture and Environmental Planning, College of Engineering and Petroleum, adhramout University, Mukalla, Yemen.
Reem Ahmed
Department of Building, Civil, and Environmental Engineering, Concordia University, Montreal, QC, Canada.
View Book :- https://stm.bookpi.org/CASTR-V6/article/view/1721
No comments:
Post a Comment