Study Of Short-term Wind Power Forcasting Based On Relevance Vector Machine

In recent years,Chinese wind power industry has developed rapidly,in order to improve the competitiveness of wind power,reduce the impact of wind power on the power network,high precision wiind power prediction is essential.In a wind farm,the high precision of short-term wind power prediction is to optimize the power quality of the fan,an important means to arrange maintenance.For power grid,reduce standby rotation capacity,reduce the wind power grid reliability problem of short-term wind power prediction is very important.This paper adopts a model of wind power prediction based on relevance vector machine for short-term wind power prediction.Kernel function type and parameters in the model to predict the influence of different was studied,and the selection of kernel function parameters using Particle Swarm Optimization Algorithm Optimization.For PSO algorithm involves population size and speed coefficient,different selection strategy of inertia weight impact on the optimization results are analyzed in the simulation.For wind power and wind speed of the nonlinear and non-stationary,the Wavelet Transform in the application and forecast,this paper studied the short-term prediction of wind power generating capacity by means of wavelet transform and relevance vector machine.Firstly,the original wind speed sequence is decomposed into a series of sub-sequences,and extracts one low frequency component and three high frequency wave components in wind speed.Secondly,combined with wind direction and temperature,these sub-sequences are respectively forecasted by RVM;Finally,respective outputs are superposed to obtain final forecasted wind power.Taking one set of measured data in a certain wind farm as application cases,the application results show that by use of wavelet transform the variation of wind speed can be grasped better,due to the better learning ability of RVM,the combination of wavelet analysis with RVM method can effectively improve the prediction accuracy,especially in wind speed mutation,this methods is superior to existing wind speed forecasting methods.