Short-Term Load Forecasting Based On Chaotic Characteristic Of Loads And Least Squares Support Vector Machines

Electric short-term load forecasting is a quite important work in the Power System. Since Power System is a huge system having a dynamic behaviour, with the development making it complicated day by day, especially with the marketization going into deep, the factors effecting load become various. To research short-term load forecasting methods adapted to the characteristic and development of electric load is an attentional question in the field.Based on the chaotic characteristic of time series of power loads and combining the phase space reconstruction theory of chaotic time series and regression theory of supporting vector machines (SVM), a short-term load forecasting model based on chaotic characteristic of loads and least squares SVM (LS-SVM) is built. At first, the phase space reconstruction of original load data is performed to form phase point series; then the phase points most adjacent to current phase points are chosen as the training samples for the proposed load forecasting model; after the decision function is found by training, the phase points involving the forecasted point can be solved; finally, reverting this phase point, the forecasted load value is obtained. Comparing the forecasting resluts by the proposed method with those from BP neural network method, the advantage and effectiveness of the proposed model in short-term load forecasting is proved.