Application of Improved Grey Theory Prediction Model in Medium-Term Load Forecasting of Distribution Network

Abstract

The progress of energy storage technology has promoted the application of distributed generation in distribution network and also increased the difficulty of load forecasting. Firstly, the paper analyze many core load impact factors such as distributed power, electric vehicles, traditional loads and so on. Considering the characteristics of the traditional grey theory prediction model with few input variables, a multivariable grey theory prediction model is proposed, the new model correct the multivariable residual error that highly improved prediction accuracy. Then the prediction model is verified by the historical load data in a certain area. The results show that the prediction accuracy of the improved prediction model is one times higher than that of the traditional prediction model. Finally, the paper uses the improved prediction model to predict the power consumption and maximum load of 2018~2020.