An Artificial Intelligent Method of Power Load Forecasting in Short-term

Due to the randomness of power load, single forecasting method is difficult to meet the requirements of forecasting accuracy. Study results show that the combination forecasting method can improve the prediction accuracy to a certain extent. This paper presents a short-term power load forecasting method based on ensemble empirical mode decomposition (EEMD), long short term memory (LSTM) and Markov chain (MC). Firstly, the original power load series data are decomposed into a series of subsequences according to different frequencies by EEMD. Secondly, the low-frequency subsequence is used as the input of multiple linear regression model to predict, the high-frequency subsequence is used as the input of LSTM to predict, and MC is used to eliminate the random error to improve the prediction accuracy. Finally, the sub forecasting results are combined to get the final load forecasting results. This method can effectively predict the change trend of each frequency component, and at the same time, it also takes into account the randomness of prediction error. The experimental results show that the combination method improves the accuracy of short-term load forecasting to a certain extent, and meets the actual engineering requirements.