Power Load Forecasting Considering Climate Factors Based on IPSO-Elman Method in China

In order to implement the national need for optimal allocation of power resources, load forecasting, as an important research topic, has theoretical and practical significance. The purpose this study is construct a prediction model considering climate factors based on large amount historical data, prove that accuracy related both regularity. results forecasting are affected by many factors, so firstly variables affecting screened. Secondly, IPSO-Elman network learning algorithm constructed taking difference between predicted value neural actual fitness function particle swarm optimization. view great influence weights thresholds Elman network, optimization (PSO) used optimize parameters in improve ELMAN network. Thirdly, with without compared analyzed, using cosine distance various error indicators. Finally, stability discriminant index regularity introduced forecast area. method proposed paper can provide reference system scheduling.