The Implementation of Support Vector Machine Method with Genetic Algorithm in Predicting Energy Consumption for Reinforced Concrete Buildings

Abstract

Accurate information on energy consumption is crucial for measuring energy efficiency and savings in buildings. It refers to the energy needed to power a building at a specific time. Energy savings can reduce costs and environmental impact by lowering greenhouse gas emissions. Obtaining precise energy consumption data is essential for all parties involved in building planning, construction, and management. Over the past decades, global energy consumption in buildings has consistently increased, with HVAC systems being a significant contributor. To tackle this problem, research developed a support vector machine model with genetic algorithms to accurately predict energy consumption in buildings. Two models were tested: a standard support vector machine and a genetic algorithm-integrated support vector machine. The test results revealed that the support vector machine model achieved an RMSE value of 2.6. Additionally, the genetic algorithm optimized the parameter C and selected the most relevant predictor variables, reducing the RMSE to 1.7 and utilizing only 3 predictor variables. In the subsequent stage, parameter optimization and function selection were performed to achieve an improved RMSE value of 1.537. This research aims to enhance energy consumption prediction for reinforced concrete buildings by combining SVM and Genetic Algorithm. SVM serves as the primary prediction model, while the Genetic Algorithm is employed to determine optimal SVM parameters and relevant features. Recent studies have demonstrated that this combination yields more accurate predictions compared to standard methods. It enables more efficient energy planning, reduced operational costs, and optimized resource utilization in reinforced concrete buildings. However, it's worth noting that this implementation may require substantial processing and resource utilization, depending on the dataset's size and complexity.

GS Cited Analysis