A Novel Method for Short-Term Wind Speed Forecasting Based on UPQPSO-LSSVM

In order to improve the accuracy of the short-term wind speed forecasting, this paper presents a novel wind speed forecasting model based on least square support vector machine (LSSVM) optimized by an improved Quantum-behaved Particle Swarm Optimization algorithm called up-weightedQPSO (UPQPSO), which uses a non-linearly decreasing weight parameter to render the importance of particles in population in order to have a better balance between the global and local searching. The developed method is examined by a set of wind speeds measured at mean half an hour of two windmill farms located in Shandong province and Hebei province, simulation results indicate UPQPSO-LSSVM model yields better predictions compared with QPSO-LSSVM and ARIMA model both in prediction accuracy and computing speed.