基于VMD-WK-OPLS的短期风速动态预测  

Dynamic Prediction of Short-Term Wind Speed Based on VMD-WK-OPLS

作　　者：朱清智[1] 董泽[2] ZHU Qing-zhi;DONG Ze(School of Automation Engineering,He Nan Polytechnic Institute,Nanyang Henan 473000,China;School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]河南工业职业技术学院自动化工程学院,河南南阳473000 [2]华北电力大学控制与计算机工程学院,北京102206

出　　处：《计算机仿真》2024年第9期84-89,309,共7页Computer Simulation

基　　金：国家自然科学基金(71471060);河南省科技攻关(232102320208)。

摘　　要：针对短期风速存在随机性、不稳定性难以建立准确预测模型,提出变分模态分解(VMD)和小波核潜在结构正交投影(WK-OPLS)的VMD-WK-OPLS预测模型,实现动态短期风速估计。首先利用VMD将风速序列分解为不同限带内的子模态,以降低短期风速非平稳性对预测结果的影响,然后对子模态分别建立WK-OPLS预测模型,并利用时滞差分(DTD)的策略对各模型动态更新,最后将各子模态预测结果组合得到最终短期风速预测值。通过宁夏风场风速序列进行验证,结果表明,本文所提出的VMD-WK-OPLS模型与EEMD-IDE-LSSVM、VMD-IDE-LSSVM、EEMD-WK-OPLS、VMD-WK-OPLS等三种模型相对比,在短期风速预测精度上得到显著提高。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 decomposition(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 prediction 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.

关 键 词：变分模态分解 时滞差分 集合经验模态分解 偏最小二乘 小波核 

分 类 号：TB99[一般工业技术—计量学]