基于EEMD-PSO-ELM的风电功率超短期预测  

Super short-term prediction of wind power based on EEMD-PSO-ELM

作　　者：毛元 冯洋 严岩 陈磊 钱勇 MAO Yuan;FENG Yang;YAN Yan;CHEN Lei;QIAN Yong(Jiangbei New Area Power Supply Company,State Grid Nanjing Electric Power Co.,Ltd.,Nanjing Jiangsu 211800,China;Training Center,State Grid Ningxia Electric Power Co.,Ltd.,Yinchuan Ningxia 750011,China;Electric Power Research Institute,State Grid Ningxia Electric Power Co.,Ltd.,Yinchuan Ningxia 750011,China)

机构地区：[1]国网南京市江北新区供电公司,江苏南京211800 [2]国网宁夏电力有限公司培训中心,宁夏银川750011 [3]国网宁夏电力有限公司电力科学研究院,宁夏银川750011

出　　处：《宁夏电力》2024年第2期1-5,26,共6页Ningxia Electric Power

摘　　要：针对风电场功率不稳定特性引起风电功率预测精度不高的问题,提出1种基于EEMD-PSO-ELM的超短期风电功率预测方法。首先,采用集合经验模态分解(ensemble empirical mode decomposition,EEMD)将风电功率序列分解为若干个模态,从而避免了模态混叠;其次,利用相空间重构对分解得到的模态计算Hurst指数,并依据Hurst指数得到最优子序列;最后,采用粒子群算法(particle swarm optimization,PSO)-极限学习机(extreme learning machine,ELM)模型对最优子序列风电功率进行预测。以某风电场为例,采用预测模型进行分析,实验结果表明EEMD-PSO-ELM预测模型的风电功率预测精度更高。Addressing the problem of low wind power prediction accuracy caused by the unstable characteristics of wind farm power,a super-short-term wind power prediction method based on ensemble empirical mode decomposition(EEMD),particle swarm optimization(PSO),and extreme learning machine(ELM)is proposed.Firstly,the wind power sequence is decomposed into several modes using EEMD to avoid mode aliasing.Secondly,phase space reconstruction is used to calculate the Hurst exponent for the decomposed modes,and the optimal sub-sequence is obtained according to the Hurst exponent.Finally,the PSO-ELM model predicts the wind power for the optimal sub-sequence.Experimental results from a specific wind farm illustrate that the EEMD-PSO-ELM prediction model achieves higher accuracy in wind power forecasting.

关 键 词：风电场功率 集合经验模态分解 相空间重构 超短期 预测精度 

分 类 号：TM615[电气工程—电力系统及自动化]