Short-term electricity load forecasting based on KPCA-Kmeans-BiGRU modeling

Aiming at the problems of too high input dimension and low prediction accuracy of existing neural network models in short-term power load forecasting, this paper proposes a short-term load forecasting method that combines kernel principal component analysis, Kmeans clustering and bidirectional gated loop unit. Firstly, the kernel principal component analysis is used to reduce the dimensionality of the historical input data, then the K-means clustering method is used to perform similar day clustering, and finally the two-way gated recurrent neural network model is used to forecast the power load. The results show that the constructed model has a goodness of fit of 0.9979, which effectively improves the prediction accuracy and is effective in short-term load forecasting.