RESEARCH ON POWER LOAD FORECASTING BASED ON SUPPORT VECTOR MACHINE

Power load forecasting is an important basic work, which is based on the past and present value of electric power load and the related factors. Load forecasting is the precondition of safe and economic operation of power system. In this paper, the basic principle of support vector machine is analysed, and the short-term load forecasting model based on support vector machine is established. A new method of power system short-term load forecasting based on support vector machine is proposed to improve the forecasting accuracy and timeliness. The application effect of the algorithm for different types of power load forecasting is also discussed. The empirical analysis shows that the load forecasting method based on support vector machine has good application effect in the short term load forecasting, the forecasting error is small and the accuracy is higher.