Energy consumption prediction of chillers based on data mining

In order to make full use of the actual operation data of chillers in energy station to improve the accuracy of energy consumption prediction, presents an energy consumption prediction model based on data mining algorithm. The model consists of three main steps: data preprocessing, modeling and analysis, and result presentation. Selects three kinds of algorithms of support vector machine, radial basis function neural network and decision tree for modeling and comparison. The results show that the energy consumption prediction model based on data mining has good practicability and reliability. Compared with the other two models, the root mean square error of the radial basis function neural network model is reduced by 0.661 and the correlation coefficient is 0.999. The radial basis function neural network has the highest accuracy of energy consumption prediction and the best modeling effect.