Power System Short-Term Load Forecasting Based on Wavelet Transform and Support Vector Machine
|School||North China Electric Power University (Hebei)|
|Course||Proceedings of the|
|Keywords||Power system Short-term load forecasting Wavelet Analysis Support Vector Machine|
Firstly, the composition of the power system load , features, and advantages and disadvantages of commonly used short - term load forecasting method are discussed. Analysis, we found that the load sequence can be seen as a superposition of a series of components having different frequency characteristics , and therefore the introduction of wavelet analysis method forecasting, the load sequence is decomposed into a sequence having different frequency characteristics , the decomposition of the component , according to its characteristics construct the corresponding support vector machine model to predict and consider the impact of factors determine the load size , temperature of meteorological factors . Finally , the prediction results of each component superimposed to obtain a final prediction value . This method is applied to a history of the area load forecasting , it can be seen that the method has a strong adaptability and better accuracy in load forecasting has significant advantages compared with other methods .