基于最大功率点跟踪的分布式光伏电量异常辨识研究
Research on Distributed Photovoltaic Electricity Abnormal Recognition Based on Maximum Power Point Tracking
翟佳 1胡晨同 1张翃帆1
作者信息
- 1. 国网北京市电力公司客户服务中心,北京 100062
- 折叠
摘要
在辨识分布式光伏电量异常时,其参数识别特征不稳定,缺少统一的可识别的辨识特征,导致不同电量模式下光伏电量异常辨识结果与实际值相差大,异常数据辨识查全率低.提出一种基于最大功率点跟踪的分布式光伏电量异常辨识方法.分析分布式光伏电池特性,采用最大功率点跟踪算法,获取与跟踪分布式光伏电量的最大功率点,利用最小二乘法对分布式光伏电量的实测值进行拟合,获取虚增电量检测指标;将虚增电量检测指标与最大功率点进行对比,分辨分布式光伏电量是否存在异常,以达到辨识光伏电量异常的目地.实验结果表明:该方法的异常辨识结果与实际值相差小、异常数据辨识查全率高,实用性强.
Abstract
The parameter identification characteristics of distributed photovoltaic power anomaly identification are unstable and lacking of unified identifiable characteristics,resulting in a large difference between photovoltaic power anomaly identification results and actual values under different power modes,and a low recall rate of abnormal data identification.To this end,this paper proposes a distributed photovoltaic power anomaly identification method based on maximum power point tracking.The characteristics of distributed photovoltaic cells are analyzed.The maximum power point tracking algorithm is used to obtain and track the maximum power point of distributed photovoltaic power.The measured value of distributed photovoltaic power is fitted by the least square method to obtain the detection index of virtual power increase.The virtual power increase detection index is compared with the maximum power point to distinguish whether there is abnormality in the distributed photovoltaic power,so as to achieve the goal of identifying the abnormality of photovoltaic power.The experimental results show that the anomaly identification results of the propose method have a small difference from the actual values,a high recall rate for anomaly data identification,and strong practicality.
关键词
最大功率点跟踪/分布式光伏/电量异常辨识/最小二乘法/虚增电量检测指标Key words
maximum power point tracking/distributed photovoltaic/electricity anomaly identification/least square method/virtual power increase detection index引用本文复制引用
基金项目
国家电网公司基金(620232220004)
国家电网北京客服中心营销稽查风险辩识防控项目(520224210004)
出版年
2024
NETL
NSTL
万方数据