基于CPC-KWS算法和混合分解方法的短期电力负荷预测OA

Power load forecasting is one of the key tasks in stable operation and rational planning of power system,which involves the accurate estimation of power demand in a certain time in the future.In order to improve the accuracy of power load forecasting,a short-term load forecasting method based on the algorithm of density peaks clustering with K-nearest neighbors and weighted similarity(DPC-KWS)and hybrid decomposition method is proposed.Firstly,the DPC-KWS algorithm is used to cluster the load data with the same power consumption behavior into one class.Secondly,the load series after clustering are decomposed into trend component and daily fluctuation component by ensemble patch transform(EPT).Then,improve complete ensemble empirical mode decomposition with adaptive noise(ICEEMDAN)decomposed into components of different frequencies;Finally,the trend component and the components of different frequencies are extracted from the sequence by temporal convolutional networks(TCN)for short-term features,and then the long-term dependencies in the data are captured and predicted by the long short-term memory network(LSTM).Finally,the forecast results are reconstructed to get the final load forecast results.The comparison with the experimental results of existing models shows that the proposed method is accurate in load forecasting,which verifies the accuracy of the proposed method.