Short-term load forecasting based on ELM combined model

Abstract

In order to accurately predict the short-term load, a combination forecasting model based on extreme learning machine is proposed. First, variational modal technology is used to decompose the original load sequence, and the appropriate number of modal components is obtained; secondly, according to the different performance characteristics of each modal, the time series and extreme learning machine model is used for prediction, and the improved bat algorithm is used to optimize the selection of parameters in the extreme learning machine; finally, the output value of the model built by each sub-sequence is reconstructed to obtain the final prediction result. Through the measured data, the effectiveness and accuracy of the combined forecasting model proposed in this paper are verified in load forecasting.