摘要
为了进一步提高预测的准确性,本文提出了一种基于人群搜索算法-小波神经网络SOA-WNN(seekeropti.mizationalgorithm-waveletneuralnetwork)的光伏短期输出功率预测算法,利用SOA在速度及全局搜索上的优势对WNN进行改进,使WNN中权值与小波因子等参数得到优化。通过与传统的WNN预测方法以及遗传算法优化的WNN预测算法进行比较,结果显示所提方法有效地提高了光伏短期输出功率预测的稳定性与准确性,具有较高的实用价值。
To further improve the forecasting accuracy,a forecasting method for photovoltaic short-term output power based on seeker optimization algorithm-wavelet neural network(SOA-WNN)is proposed.By taking advantages of SOA, such as high speed and global search,WNN is modified to optimize its parameters including weight and wavelet factor. Compared with the traditional WNN forecasting method and the WNN forecasting methodoptimized bygenetic algo. rithm,the proposed method effectively improves the stability and accuracy of photovoltaic short-term power output fore. casting,indicating that it has higher practical values.
作者
高毅
李盛伟
迟福建
葛磊蛟
张东
GAO Yi;LI Shengwei;CHI Fujian;GE Leijiao;ZHANG Dong(Economic and Technical Research Institute,State Grid Tianjin Electric Power Company,Tianjin 300171,China;State Grid Tianjin Electric Power Company,Tianjin 300055,China;School of Electrical andInformation Engineering,Tianjin University,Tianjin 300072,China)
出处
《电力系统及其自动化学报》
CSCD
北大核心
2019年第6期62-66,共5页
Proceedings of the CSU-EPSA
基金
国网电网公司科技资助项目(KJ17-1-06)
关键词
人群搜索算法
光伏输出功率
小波神经网络
优化
seeker optimization algorithm(SOA)
photovoltaic output power
wavelet neural network(WNN)
optimi-zation
