A Circular Recursion Algorithm Power Load Forecasting Model and Application

Abstract

Load forecasting is the basis and premise of safe and reliable economic operation and planning of power systems. This paper analyzes the advantages and disadvantages of various load forecasting models, including gray model (GM), support vector machine (SVM), back propagation (BP), neural network method (NNM), combination forecasting method (CFM) and expert system method (ESM). Based on the analysis of the traditional power load forecasting algorithm of the power grid, according to the characteristics of different forecasting time phases, a boolean recursive algorithm (BRA) for power load characteristic prediction is proposed. In this model, the whole year is divided into five feature prediction stages, the prediction algorithm formulas of different feature stages are different. Based on the new algorithm, a set of load forecasting software are designed. This paper introduces the main functions and features of the software and describes how the software is predicted. And the BRA uses the actual historical data values of the power supply company to predict the power load, the application shows the feasibility.