Fuzzy modeling technique with PSO algorithm for short-term load forecasting

This paper proposes a new modeling approach for building TSK models for short-term load forecasting (STLF). The approach is a two-stage model building technique, where both premise and consequent identification are simultaneously performed. The fuzzy C-regression method (FCRM) is employed at stage-1 to identify the structure of the model. The resulting model is reduced in complexity by selection of the proper model inputs which are achieved using a Particle Swarm Optimization algorithm (PSO) based selection mechanism at stage-2. To obtain simple and efficient models we employ two descriptions for the load curves (LC's), namely, the feature description for the premise part and the cubic B-spline curve for the consequent part of the rules. The proposed model is tested using practical data, while load forecasts with satisfying accuracy are reported.