Short-term load Forecasting Method Based on Load regularity Analysis for Supporting Vector Machines

A short-term load forecasting method in power system based on discrete fourier transform and support vector machines is proposed in this essay.Historical load data is decomposed into the series with different frequency characteristics by discrete fourier transform method.According to the features of load data,different frequency series are composed of four components.Different models are constructed to deal with the different components.Support vector machines is used to deal with the component which is influenced most by temperature.The feasibility and validity of the method in this paper is proved by practical example.