Building Cooling Load Forecasting Model Based on LS-SVM

Abstract

A number of different forecasting methods have been proposed for cooling load forecasting including historic method, real-time method, time series analysis, and artificial neural networks (ANN), but accuracy and time efficiency in prediction are a couple of contradictions to be hard to resolve for real-time traffic information prediction. In order to improve time efficiency of prediction, a new hourly cooling load prediction model and method based on Least Square Support Vector Machine (LS-SVM) is proposed in this paper. A comparison of the performance of LSSVM with back propagation neural network (BPNN) is carried out. Experiments results demonstrate that LSSVM can achieve better accuracy and generalization than the BPNN, the LSSVM predictor can reduce significantly both relative mean errors and root mean squared errors of cooling load.