一种基于VMD-PSO-SVM的短期风电功率预测算法

A Design of PSO-SVM Short-term Wind Power Forecasting Algorithm Based on VMD

基于风场数据,利用风电机组功率特性曲线,采用支持向量机(Support Vector Ma?chine,SVM)非线性拟合方法,设计了一种基于变分模态分解(Variational Model Decomposi?tion,VMD)的粒子群优化支持向量机短期功率预测算法.首先,通过VMD将原始风功率序列分解为多组平稳的固有模态函数和趋势项,对风电功率数据进行预处理;其次,利用粒子群算法优化(Particle Swurm Optimization,PSO)支持向量机参数,建立VMD-PSO-SVM组合预测算法模型,对每组固有模态函数和趋势项进行预测,得到多组预测结果,再将其重组,得到功率预测结果;最后,将预测结果与其他预测算法进行对比,结果表明预测精度更高.

In order to improve the accuracy of wind power prediction,combined with wind turbine power characteristic curve and support vector machine nonlinear fitting,a short-term power prediction algorithm based on Particle Swarm Optimization(PSO)Support Vector Machine(SVM)is designed.Firstly,the original wind power series is decomposed into several groups of stationary intrinsic mode functions and trend terms by Variational Model Decomposition(VMD),so as to realize the preprocessing of wind power data.Secondly,particle swarm optimization is used to optimize the parameters of Support Vector Machine(SVM),and a combined prediction algorithm model VMD-PSO-SVM is established.Then,the natural mode functions and trend terms of each group are predicted,and several groups of prediction results are obtained.Then,these results are recombined to get the power prediction results.Finally,the prediction results of VMD-PSO-SVM are compared with those of traditional PSO-SVM,which shows that the accuracy is higher and in line with the actual needs.