Support Vector Machine Model in Electricity Load Forecasting

Abstract

With the development of electronic industry, accurate load forecasting of the future electricity demand plays an important role in regional or national power system strategy management. Electricity load forecasting is difficult due to the nonlinearity of its influencing factors. Support vector machine (SVM) have been successfully applied to solve nonlinear regression and time series problems. However, the application to load forecasting is rare. In this study, a model of support vector machine is proposed to forecast electricity load. The model overcomes the disadvantages of general artificial neural network (ANN), such as it is not easy to converge, liable to trap in partial minimum and unable to optimize globally, and the generalization of the model is not good, etc. The SVM model ensured the forecasting is optimized globally. Subsequently, examples of electricity load data from Hebei province of China are used to illustrate the performance of the proposed model. The empirical results reveal that the proposed model outperforms the general artificial neural network model, and the forecasting accuracy improved effectively. Therefore, the model provides a promising arithmetic to forecasting electricity load in power industry