Energy Consumption Pattern Analysis in a Large Public Library of China

This study analyzes energy consumption in a large public library in China using regression and machine learning models. Data from 238 circuits revealed key usage patterns, including midnight surges and dual peaks, along with anomalies like unexpected energy spikes. Model comparisons identified temperature, humidity, and time of day as major influencing factors. The study recommends optimizing HVAC and mechanical circuit operations to improve efficiency. Findings contribute to reducing operational costs and carbon emissions, offering insights for energy management in public institutions. The methodologies can be applied to other buildings seeking sustainability and efficiency improvements.