Research on Energy Consumption and Carbon Emission Prediction for Enterprises in Dongguan Based on Machine Learning

With the intensification of global climate change, the management of energy consumption and carbon emissions by enterprises has become increasingly crucial. As a major manufacturing center in China, Dongguan faces significant pressures on energy and carbon emissions in its economic development. This study develops a machine learning model combining random forest regression and polynomial regression to accurately predict carbon emissions of enterprises in Dongguan. The model leverages the capability of random forest regression to effectively capture nonlinear relationships, while polynomial regression enhances trend validation, resulting in an overall prediction accuracy exceeding 92% for enterprise carbon emissions. Through this model, we have provided an in-depth analysis of the coupling relationship between electricity consumption and carbon emissions, offering a scientific basis for enterprises to develop effective carbon reduction strategies and promoting sustainable development.