Short-Term Load Forecasting Using Support Vector Machine with SCE-UA Algorithm

Short-term load forecasting (STLF) has always been a very important issue in power system planning and operation. Recently, along with power system privatization and deregulation, accurate forecast of electricity load has received increasing attention. However, forecasting electricity load is difficult because of the randomness and uncertainties of load demand. Support vector machine (SVM) is a novel type of learning machine, which has been successfully employed to solve nonlinear regression and time series problems and showed its potential in STLF. However, the accuracy of STLF is greatly related to the selected parameters of SVM. In this paper, the SCE-UA algorithm, which is an effective and efficient method to optimize model parameters and widely applied to optimize the parameters of the hydrologic models, is employed to optimize the parameters of a SVM model. Subsequently, examples of electricity load data from GuiZhou Power Grid, China were used to compare the forecast performance of the proposed SCE-UA SVM model and back propagation neural network(BPNN) which has obtained wide attention in STLF . The results reveal that the proposed model outperforms the BPNN model. Consequently, the proposed SCE-UA SVM model provides a promising alternative for forecasting electricity load.