Application of a Load Forecasting Model Based on Improved Grey Neural Network in the Smart Grid

Abstract

An important feature of smart grid is the intelligent power distribution function based on load forecasting with high accuracy. Accurate prediction of load is the key indicator of power intelligence. As a result of this, this paper combines Genetic Algorithm with grey prediction model, uses Genetic Algorithm to optimize the initial value and the background value of traditional GM (Grey Model), combines the new GM with BP neural network and constructs a tandem Grey Neural Network model, which is used in load forecasting in smart grid. This model can solve the forecasting problem of non-isometric series, greatly improve the accuracy of prediction model, optimize data quality, strengthen the intelligence on operation and deployment, and provide more realistic, workable scientific reference for the decision support of smart grid. Finally the proposed method is applied to predict the load of some area. The results prove the effectiveness of the method.