摘要
针对风电功率预测中出现的随机性和波动性问题,提出了一种基于变分模态分解(VMD)和JAYA优化最小二乘支持向量机(LSSVM)参数的算法,实现短期风电功率的预测。该算法通过分析历史风速序列和气压对风电功率的影响,用VMD对历史风速进行分解,将分解出的风速分量结合气象因素中的气压作为LSSVM预测模型的训练输入,利用JAYA算法的寻优特性对LSSVM的参数进行优化,建立短期风电功率预测模型。最后以风电场实测数据为例进行仿真分析,仿真结果表明,与LSSVM和PSO优化的LSSVM预测模型相比,VMD-JAYA-LSSVM的方法对短期风电功率预测的精度提高了5.2%。
To solve the problems of randomness and volatility in wind power prediction,an algorithm based on variational mode decomposition(VMD)and JAYA optimization of least squares support vector machine(LSSVM)parameters is proposed to realize short-term wind power prediction.By analyzing the influence of historical wind speed sequence and atmospheric pressure on wind power,the algorithm decomposes the historical wind speed with VMD,takes the decomposed wind speed components combined with the atmospheric pressure in meteorological factors as the training input of LSSVM prediction model,and uses the optimization characteristics of JAYA algorithm to optimize the parameters of LSSVM and establish a short-term wind power prediction model.Finally,the simulation analysis is carried out with the measured data of the wind farm as an example.The results show that compared with LSSVM and LSSVM optimized by particle swarm optimization(PSO),VMD-JAYA-LSSVM improves the accuracy of short-term wind power prediction by 5.2%.
作者
陶凯
吴定会
TAO Kai;WU Ding-hui(College of Internet of Things Engineering,Jiangnan University,Wuxi 214122 China)
出处
《控制工程》
CSCD
北大核心
2021年第6期1143-1149,共7页
Control Engineering of China
基金
国家自然科学基金资助项目(61572237)。
