Energy Saving Prediction Method for Public Buildings Based on Data Mining

The energy consumption monitoring system of public buildings has accumulated a lot of data in the actual operation process, which needs to be fully analyzed to get valuable information and relevance information. In this paper, related research, simulation and verification are used to obtain the distribution law of regional energy consumption influencing factors and energy consumption parameters, and the energy-saving transformation prediction of energy consumption stochastic model is proposed. At the same time, with the help of the idea of reverse modeling, a prediction stochastic model of energy-saving transformation based on data mining is established. In this scheme, genetic algorithm is also used to optimize the sub tree generation process of gradient lifting decision tree, and the problem of C4.5 decision tree in short-term prediction is improved, to achieve energy consumption prediction. The experimental results show that the prediction accuracy of the model is higher than that of the traditional regression model.