基于传递闭包的光伏短期功率组合预测方法研究  

RESEARCH OF COMBINED FORECASTING METHOD OF SHORT-TERM PHOTOVOLTAIC POWER ON TRANSITIVE CLOSURE BASED

作　　者：李薇 王鑫鹏 许野 王旭 马文静 Li Wei;Wang Xinpeng;Xu Ye;Wang Xu;Ma Wenjing(College of Environmental Science and Engineering,North China Electric Power University,Beijing 102206,China;China National Energy Group Technical and Economic Research Institute,Beijing 102211,China)

机构地区：[1]华北电力大学环境科学与工程学院,北京102206 [2]国家能源集团技术经济研究院,北京102211

出　　处：《太阳能学报》2023年第6期265-274,共10页Acta Energiae Solaris Sinica

基　　金：国家重点研发计划(2018YFE0208400);国家电网有限公司总部科技项目《面向跨境互联的多能互补新型能源系统关键技术研究》。

摘　　要：为有效提升光伏短期发电功率的预测精度和保障电网的稳定运行,在分别利用皮尔逊相关系数法(PCC)提取相关气象要素、传递闭包法确定相似日的基础上,结合ARIMA时间序列和BP神经网络,分别构建PCC-传递闭包-ARIMA和PCC-传递闭包-BP两类组合预测模型,用于解决云南宾川某光伏发电站的出力预测问题。对比结果显示:(1)相较于单一模型,组合预测模型的预测精度明显提升,其中PCC-传递闭包-BP模型的预测效果最好,平均精度可达91.19%;(2)在历史光伏出力普遍偏高的前提下,ARIMA模型在高光伏出力时段可较好地描述光伏出力的波动和变化特性,BP模型在低光伏出力时段具有更强的调整和纠正能力。In order to effectively improve the prediction accuracy of short-term photovoltaic power generation and ensure the stable operation of the power grid,this study uses the Pearson Correlation Coefficient(PCC)method to extract relevant meteorological elements and adopts the transitive closure method to determine the similar days.Combining ARIMA time series and BP neural network,two combined prediction models of PCC-transitive closure-ARIMA and PCC-transitive closure-BP are constructed,and used to solve the output prediction problem of a photovoltaic power station in Binchuan,Yunnan.The comparison results demonstrate that,(i)compared with single model,the prediction accuracy of two combination models is improved significantly,where the combined model based on BP neural network has the better performance,with an average prediction accuracy of 91.19%;(ii)ARIMA model is suitable to describe the fluctuation and variation characteristics of photovoltaic output in the period of high-level electricity output;correspondingly,BP model owns more adjust and correct capability under unfavorable meteorological condition.

关 键 词：光伏发电 预测模型 因子分析 皮尔逊相关系数 传递闭包 ARIMA BP神经网络 

分 类 号：TM615[电气工程—电力系统及自动化]