Predictive Modeling for Building Energy Consumption in Python

Abstract

The increasing demand for energy efficiency in commercial and residential buildings necessitates the development of accurate predictive models for energy consumption. This thesis presents a comprehensive approach to developing and simulating a predictive model for building energy consumption using a multiple linear regression algorithm implemented in Python. The model utilizes real-world data including temperature, humidity, and energy consumption from different floors to forecast total building energy usage. The study leverages Python’s powerful data analysis capabilities to derive model coefficients and integrates them into a dynamic live model, allowing for real-time simulation and analysis of energy flow. The results demonstrate that the model accurately predicts energy consumption. The developed model achieved a Root Mean Square Error (RMSE) of 2.84, a Mean Absolute Percentage Error (MAPE) of 4.72%, and a coefficient of determination (R²) of 0.91, indicating strong predictive performance and reliability. The model provides a reliable tool for building energy managers to implement proactive load control strategies and optimize energy usage. This work validates the efficacy of using a combined Python programming and data analysis approach for predictive modeling in smart building management systems.