基于OS-ELM的光伏发电中长期功率预测  

Medium and Long Term Photovoltaic Power Generation Forecasting Based on OS-ELM

作　　者：钱子伟 孙毅超 王琦[1,2] 季顺祥 周敏[1,2] 曾柏琛 Qian Ziwei;Sun Yichao;Wang Qi;Ji Shunxiang;Zhou Min;Zeng Baichen(School of NARI Electrical and Automation,Nanjing Normal University,Nanjing 210023,China;Jiangsu Key Laboratory of Gas and Electricity Interconnection Integrated Energy,Nanjing Normal University,Nanjing 210023,China)

机构地区：[1]南京师范大学南瑞电气与自动化学院,江苏南京210023 [2]南京师范大学江苏省电气互联综合能源工程实验室,江苏南京210023

出　　处：《南京师范大学学报（工程技术版）》2020年第1期8-14,共7页Journal of Nanjing Normal University(Engineering and Technology Edition)

基　　金：江苏省研究生科研与实践创新计划项目(SJCX19_0386)

摘　　要：为了进一步提高光伏出力预测的精度,提出了一种基于在线序列极限学习机的光伏发电中长期功率预测方法.结合在线序列极限学习机学习速度快、泛化能力强的特点,通过对大量气象数据和历史发电数据综合处理,对光伏发电系统的输出功率进行预测.同时,由于实时数据的不断输入,该方法能够对预测模型进行在线更新.算例仿真研究表明,该预测方法与反向传播神经网络、支持向量机方法相比,能够有效提高预测精度,满足在线应用的需求,具有较好的应用前景.In order to further improve the accuracy of PV output prediction,a medium and long term power prediction method based on online sequential extreme learning machine(OS-ELM)is proposed.Combined with the characteristics of fast learning and generalization ability of OS-ELM,the output power of photovoltaic power generation system is predicted by comprehensively processing a large number of meteorological data and historical power generation data.At the same time,due to the continuous input of real-time data,the method can update the prediction model online.The simulation study shows that compared with the back propagation(BP)neural network and support vector machine(SVM)method,the predictional method can effectively improve the prediction accuracy and meet the needs of online applications,and it has a good application prospect.

关 键 词：光伏预测 相关性分析 在线序列极限学习机 数据更新 

分 类 号：TM615[电气工程—电力系统及自动化]