Power Load Forecasting Based on Improved Genetic AlgorithmGM(1,1) Model
Dong-xiao Niu,Wei Li,Zhu-hua Han,Xiu-e Yuan
DOI: https://doi.org/10.1109/ICNC.2008.836
2008-01-01
Abstract:According to Traditional Grey Model (GM(1, 1)) is not accurate and the value of parameter is constant, in order to overcome these disadvantages, this paper put forward an improved genetic algorithm-GM(1, 1) (IGA-GM (1, 1)) to solve the problem of short-term load forecasting (STLF) in power system. The proposed algorithm not only improved the original series but also constructed optimal grey model GM(1, 1) to enhance the accuracy of forecasting, and the improved decimal-code genetic algorithm (GA) is applied to search the optimal value of grey model GM(1, 1). What's more, this paper also proposes the one-point linearity arithmetical crossover in genetic algorithm, which can greatly improve the speed of crossover and mutation. At last, this proposed algorithm improved the residual error test which lead to the results more accurate, and a comparison of the performance has been made between IGA-GM(1, 1) and traditional GM(1, 1) forecasting model. Results show that the IGA-GM(1, 1) had better accuracy and practicality.