The short-term load forecasting of electric power system based on combination forecast model

Abstract

Because of the power system load forecasting in the constant weight combination forecast method when there is a single method mutation of prediction results, affect the prediction precision. Therefore, this paper proposes an improved combination forecast method, the least squares algorithm, Kalman filter algorithm, chaos Kalman filtering algorithm further combination, taking the variable weights method to modification the model, for electric power system short-term load forecasting. Through the simulation analysis to comparing improved the combination forecast method with single prediction methods, and the constant weight combination forecast method (variance-covariance method, the optimal weighted method). the result shows that combined forecasting method is better than single prediction methods, and the prediction precision of improved combination forecast method is higher, the forecast effect is more ideal.