Research on Energy Alternation Based on Gaussian Regression Combined Prediction Model

At present, the problems of energy shortage and environmental damage in China are becoming more and more serious. In order to encourage enterprises to reduce energy consumption and improve environmental quality, the research on industrial energy alternatives is carried out for the needs of electric power enterprises to promote energy replacement. First, for the excellent enterprises that have made energy substitution, the Apriori algorithm is used to identify the frequent itemsets related to electric energy substitution, and the association rules are mined. Then, based on the clustered feature set, the Gaussian regression prediction model is established to estimate the expected cost and uncertainty of the enterprise energy substitution. Through the linear combination of Gaussian models, various comprehensive prediction targets are obtained. Finally, based on the principle of thermal energy equivalent, the highest advantage price, that is the boundary electricity price, is obtained, which providing data support for the electricity price. Experiments show the feasibility and accuracy of the method, so as to assist the power companies to develop user energy alternatives to provide implementation guidance.