Improved load forecasting model based on two-stage optimization of gray model with fractional order accumulation and Markov chain

Accurate prediction of load is very important in the management of power system. This paper proposed a two-stage optimization method of gray-Markov model with fractional order accumulation for short-term load forecasting. Firstly, the gray model GMpq(1,1) with the optimal fractional order accumulation is investigated. Secondly, the Markov chain model with optimal state numbers is used to deal with residual error series to improve prediction accuracy. Thirdly, the results of case study show that the proposed gray-Markov model with fractional order accumulation and two-stage optimization method can perform more intelligent and further improve prediction accuracy in the short-term load forecasting relative to some other methods.