Research on SVM Wind Power Prediction Based on Heuristic Algorithm

With the disadvantages of volatility, intermittent and randomness of wind power, a research on constructing a fairly accurate prediction model is imperative to improve the quality of power system. Considering the optimization ability of heuristic algorithm and the regression ability of support vector machine, a HA-SVM model is constructed.Case study shows that, compared with other heuristic algorithms, the search efficiency and speed of differential evolution are good, and the prediction accuracy of the model is high.