Medium and Long-Term Electric Power Planning Load Forecasting Based on Variable Weights Gray Model

摘要

Based on the gray forecasting theory, the paper studies the influence of the buffer operator on the gray forecasting results, and solves the problems of the traditional gray model in the medium and long-term electric load forecasting. By constructing the buffer operator, the dynamic preprocessing of the raw load data is realized. Based on particle swarm optimization (PSO) algorithm, taking the maximum gray correlation degree between fitted value and actual values as the goal, to choose the optimal buffer operator and buffer system, which can reasonably weak the historical load data, improve precision of the model and make a prediction results more reasonable real.