Real time prediction method of energy consumption of geothermal system in public buildings based on wavelet neural network

Abstract

Real time prediction of energy consumption is the basis of energy conservation and emission reduction. Aiming at the problems of large prediction error and poor effect, a real-time prediction method of energy consumption of geothermal system of public buildings based on wavelet neural network is proposed. Firstly, the energy consumption of geothermal system in public buildings is analyzed, the wavelet neural network is designed, the neural network is optimized and solved by genetic algorithm, and the necessity of constructing the real-time prediction model of energy consumption based on wavelet neural network is established. Then it introduces the basic principle of model establishment, wavelet analysis, and shows the role of wavelet analysis in prediction model. Finally, based on the distribution structure of public buildings, this paper analyzes the energy consumption system of geothermal system, constructs the energy consumption prediction method, analyzes the over?all temperature regulation energy consumption prediction principle of building geothermal system, and realizes the real-time prediction of energy consumption of geothermal system of public buildings. The experimental results show that the energy consumption real-time prediction results of the designed method are basically similar to the actual prediction values, and the prediction efficiency is high, which can effectively reduce the energy consumption of the geothermal system of public buildings.