CRPSO-based automatic TSK fuzzy model extraction for one hour ahead load forecasting

Electric load forecasting is essential for effective power system planning and operation because many decisions related to power system planning and operation depends on future behavior of electric loads. In this paper, we present automatic TSK fuzzy model extraction method for one hour ahead load forecasting. To extract TSK fuzzy model automatically, cooperative random learning particle swarm optimization (CRPSO) proposed by Zhao et al. is used. Structure and parameter of fuzzy model can be simultaneously extracted by CRPSO. To confirm the effectiveness of CRPSO-based method, two kinds of real world hourly load dataset are used and the performance of the CRPSO-based fuzzy model is compared with ANN and SVR. The experimental results show that CRPSObased TSK fuzzy model outperforms other methods.