Accurate Forecasting of Electric Load Demand via Interval Type-2 Fuzzy Logic Method With Particle Swarm Optimization

Since the electrical load is significantly affected by many factors, such as the economic policy, the human behaviors, the political environment, etc., it is of great difficult to achieve the accurate forecasting. In order to solve such problem, an electric load demand prediction model is proposed in this paper firstly, by utilizing the interval type-2 fuzzy logic (IT2FL) method. Then, the particle swarm optimization (PSO) algorithm is applied to obtain the best parameters of the model, including the antecedents and consequents of the designed fuzzy rules, aiming at further improving the forecasting performance. Finally, experimental and comparison are conducted. And, both of them demonstrate the superior performance of the proposed forecasting model.