Research On Power Load Combination Forecasting Based On Intelligent Optimization Method

As an important part of the electric power operation system,short-term electric load forecasting has been the focus of researchers.The historical data of power load is essentially a random non-stationary sequence,and completely error-free prediction is currently impossible.Therefore,researchers have been working to improve the accuracy of prediction.This topic mainly proposes the following two prediction methods:(1)Based on the idea of combined forecasting,a combined forecasting model based on bird swarm algorithm is proposed.Firstly,three methods are used to establish the prediction model,which are random forest,extreme learning machine and Elman neural network.Then the bird swarm algorithm is used to optimize the weights of the three models.By comparing the experimental prediction results of the combined model and the single model,it can be indicated that the combined prediction model works better.(2)In order to reduce the complexity of the load sequence,an optimized ELM prediction model based on variational mode decomposition is proposed.Firstly,the original load data is processed by the variational mode decomposition method,and the load subsequence with low relative complexity is obtained.Then the subsequence is built to predict the model,and the prediction model used is the limit learning model optimized by the e beetle swarm optimization algorithm..The beetle swarm algorithm combines the beetle swarm algorithm and the particle swarm algorithm to achieve better algorithm performance.The final prediction result is obtained by combining multiple prediction results.Through experimental comparison,the proposed prediction model is more accurate.