Dynamic Prediction of Short-Term Wind Speed Based on VMD-WK-OPLS
Aiming at the problem of randomness and instability of short-term wind speed,which makes it difficult to establish an accurate prediction model,a VMD-WK-OPLS prediction model based on variational mode decomposi-tion(VMD)and wavelet kernel potential structure orthogonal projection(WK-OPLS)is proposed to realize dynamic short-term wind speed estimation.Firstly,VMD is used to decompose the wind speed series into sub models within different band limits to reduce the impact of short-term wind speed unsteadiness on the prediction results.Then,WK-OPLS prediction models are established for sub models,and the DTD strategy is used to dynamically update each model.Finally,the prediction results of each sub model are combined to obtain the final short-term wind speed pre-diction value.The results show that the VMD-WK-OPLS model proposed in this paper has significantly improved the short-term wind speed prediction accuracy compared with EEMD-IDE-LSSVM,VMD-IDE-LSSVM,EEMD-WK-OPLS,VMD-WK-OPLS and other three models through the verification of the wind speed series of Ningxia Wind Farm.
NETL
NSTL
万方数据
