Optimization of Building Power Consumption Prediction Model

Abstract

Energy use and related greenhouse gas emissions have dramatically increased over the last century as a result of a range of factors, including both technological and population-based ones. Consequently, improving our energy efficiency is crucial if we are to attain global sustainability. Planning, management, optimization, and conservation are just a few of the many areas where predicting the energy use of a building is crucial. Due to their improved performance, robustness, and simplicity of implementation, data-driven models for energy forecasting have developed dramatically over the past few decades. Finding the right set of hyper-parameters is crucial to predict power usage using machine learning models. This raises the issue of complexity and extensive search spaces. We used the Metaheuristics approach to discover the ideal learning parameter values to address this issue. Through automatic parameter optimization, models optimized with metaheuristic methods typically perform better