基于改进K均值聚类和加权动态时间规整的分布式光伏异常数据辨识方法

摘要
图/表
参考文献
相关文章 (14)

全文: PDF (924 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要光伏发电设备故障及外界环境等多种因素导致分布式光伏在发电过程中产生大量异常数据。为了提高数据处理的准确性和效率,本文提出一种基于改进K均值聚类算法和加权动态时间规整（WDTW）的分布式光伏异常数据辨识方法。首先,对分布式光伏发电数据进行分析,利用同时段功率均值法对异常数据进行初步剔除。通过光照强度数据的归一化处理,提出基于改进K均值聚类算法的光伏数据相似日划分方法。其次,考虑光伏数据在时间维度的变化性和复杂性,引入异常数据识别最好的时段和阈值因子,提出基于WDTW的数据相似度分析方法。利用相似度计算轮廓系数,对光伏发电异常数据进行二次剔除。仿真结果表明,所提方法在辨识分布式光伏异常数据方面具有显著优势,相比于现有的四分位法、3-sigma法和特征聚类法,所提方法的辨识精度分别提高6.92%、9.00%和8.12%,同时计算复杂度降低。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	杨旺霞
	李本瑜
	翟苏巍
	石恒初
	李银银

关键词 ：改进K均值聚类算法, 加权动态时间规整（WDTW）, 分布式光伏, 异常数据辨识

Abstract：The failure of photovoltaic power generation equipment and various factors such as external environment lead to a large number of abnormal data during the power generation process. In order to improve the accuracy and efficiency of data processing, this paper proposes a distributed photovoltaic abnormal data identification method based on improved K-means algorithm and weighted dynamic time warping (WDTW). Firstly, the distributed photovoltaic power generation data is analyzed, and the abnormal data is preliminary eliminated by means of the simultaneous power mean method, and a photovoltaic data similarity day partitioning method based on improved K-means algorithm is proposed by normalizing the light intensity data. Secondly, considering the variability and complexity of photovoltaic data in the time dimension, a data similarity analysis method based on WDTW is proposed by introducing the best time period and threshold factor for identifying abnormal data. The similarity is used to calculate the contour coefficient, and the residual abnormal photovoltaic power generation data is culled twice. The simulation results show that the proposed method has significant advantages in identifying distributed photovoltaic abnormal data. Compared with the existing quartile method, 3-sigma method, and feature clustering method, the identification accuracy has been improved by 6.92%, 9.00%, and 8.12% respectively, while the computational complexity is reduced.

Key words： improved K-means clustering algorithm weighted dynamic time warping (WDTW) distributed photovoltaic identification of abnormal data

收稿日期: 2024-11-11

基金资助:云南电网公司科技项目（YNKJXM20230368）

作者简介: 杨旺霞（1981—）,女,云南省大理市人,硕士,高级工程师,主要从事电力系统继电保护研究工作。

引用本文:

杨旺霞, 李本瑜, 翟苏巍, 石恒初, 李银银. 基于改进K均值聚类和加权动态时间规整的分布式光伏异常数据辨识方法[J]. 电气技术, 2025, 26(5): 39-47. YANG Wangxia, LI Benyu, ZHAI Suwei, SHI Hengchu, LI Yinyin. A distributed photovoltaic abnormal data identification method based on improved K-means clustering algorithm and weighted dynamic time warping. Electrical Engineering, 2025, 26(5): 39-47.

链接本文:

https://dqjs.cesmedia.cn/CN/Y2025/V26/I5/39

[1] 廖家齐, 于若英, 刘瑜俊, 等. 基于自适应高斯混合模型的含高渗透率分布式光伏电力系统风险评估[J]. 电力系统保护与控制, 2024, 52(19): 144-156.
[2] 陈艳波, 刘宇翔, 田昊欣, 等. 基于广义目标级联法的多牵引变电站光伏-储能协同规划配置[J]. 电工技术学报, 2024, 39(15): 4599-4612.
[3] 武昭原, 刘婧宇, 周明, 等. 分散决策下分布式光伏储能系统外部性价值量化评估[J]. 电力系统自动化, 2024, 48(5): 38-47.
[4] 李斌, 罗晓伊. 分布式电源对电力系统电压无功优化影响的研究[J]. 电气技术, 2024, 25(10): 55-61, 78.
[5] MA Wenting, MA Mingyao, ZHANG Zhixiang, et al.Anomaly detection of mountain photovoltaic power plant based on spectral clustering[J]. IEEE Journal of Photovoltaics, 2023, 13(4): 621-631.
[6] 张铄, 吴丽珍. 计及坏数据辨识与修正的配电网状态估计[J]. 电气技术, 2022, 23(11): 1-6, 12.
[7] SU Binyi, ZHOU Zhong, CHEN Haiyong.PVEL-AD: a large-scale open-world dataset for photovoltaic cell anomaly detection[J]. IEEE Transactions on Industrial Informatics, 2023, 19(1): 404-413.
[8] 李阳, 沈小军, 张扬帆, 等. 基于速度-关联约束的风电机组风速感知异常数据识别方法[J]. 电工技术学报, 2023, 38(7): 1793-1807.
[9] 叶林, 崔宝丹, 李卓, 等. 光伏电站高比例异常运行数据组合识别方法[J]. 电力系统自动化, 2022, 46(20): 74-82.
[10] BENALCAZAR P, KOMOROWSKA A, KAMINSKI J.A GIS-based method for assessing the economics of utility-scale photovoltaic systems[J]. Applied Energy, 2024, 353: 122044.
[11] 顾菊平, 赵佳皓, 张新松, 等. 电力设备多参量监测数据清洗研究现状及展望[J]. 高电压技术, 2024, 50(8): 3403-3420.
[12] GAO Jianwei, WANG Yaping, GUO Fengjia, et al.A two-stage decision framework for GIS-based site selection of wind-photovoltaic-hybrid energy storage project using LSGDM method[J]. Renewable Energy, 2024, 222: 119912.
[13] 王丽朝, 孟子尧, 陈诗明, 等. 基于GRU神经网络的光伏电站数据预处理方法[J]. 太阳能学报, 2022, 43(11): 78-84.
[14] WANG Rui, LI Peng, YU Hao, et al.Identification of critical uncertain factors of distribution networks with high penetration of photovoltaics and electric vehicles[J]. Applied Energy, 2023, 329: 120260.
[15] JAVAID A, SHAFI I, KHALIL I U, et al.Enhancing photovoltaic systems using Gaussian process regression for parameter identification and fault detection[J]. Energy Reports, 2024, 11: 4485-4499.
[16] 郝颖, 冬雷, 王丽婕, 等. 基于数学形态学去噪的光伏发电限电异常数据识别算法[J]. 中国电机工程学报, 2022, 42(21): 7843-7854.
[17] 周嘉琪, 毕利. 基于GAN的光伏逆变器数据异常检测技术[J]. 电力系统保护与控制, 2022, 50(1): 133-140.
[18] MA Mingyao, ZHANG Zhixiang, YUN Ping, et al.Photovoltaic module current mismatch fault diagnosis based on I-V data[J]. IEEE Journal of Photovoltaics, 2021, 11(3): 779-788.
[19] BADR M M, ABDEL-KHALIK A S, HAMAD M S, et al. Intelligent fault identification strategy of photo- voltaic array based on ensemble self-training learning[J]. Solar Energy, 2023, 249: 122-138.
[20] SANTOS M F O, DE SOUZA MELO W JR, OLIVEIRA DE SA A, et al. A hybrid cyber-physical risk identification method for grid-connected photo- voltaic systems[J]. Sustainable Energy, Grids and Networks, 2024, 39: 101490.
[21] LIU Bo, CHEN Lei, SUN Kai, et al.A hot spot identification approach for photovoltaic module based on enhanced U-net with squeeze-and-excitation and VGG19[J]. IEEE Transactions on Instrumentation and Measurement, 2024, 73: 3516510.
[22] FAN Siyuan, WANG Yu, CAO Shengxian, et al.A deep residual neural network identification method for uneven dust accumulation on photovoltaic (PV) panels[J]. Energy, 2022, 239: 122302.
[23] OSORIO L, MORENO M, RIVERA M, et al.A metaheuristic-based method for photovoltaic temperature computation under tropical conditions[J]. Solar Energy, 2024, 271: 112414.
[24] ASLANI M, SEIPEL S.Automatic identification of utilizable rooftop areas in digital surface models for photovoltaics potential assessment[J]. Applied Energy, 2022, 306: 118033.
[25] SIMAO T E, VALLE B C, ROSA Y C, et al.A method for the estimation of missing strings in very- large-scale photovoltaic power plants[J]. IEEE Journal of Photovoltaics, 2024, 14(5): 839-847.
[26] CHEN Xiang, JIANG Meng, DING Kun, et al.Fault prediagnosis, type identification, and degree diagnosis method of the photovoltaic array based on the current- voltage conversion[J]. IEEE Transactions on Power Electronics, 2024, 39(12): 16708-16719.
[27] CHEN Di, PENG Qiuzhi, LU Jiating, et al.Classifi- cation and segmentation of five photovoltaic types based on instance segmentation for generating more refined photovoltaic data[J]. Applied Energy, 2024, 376: 124296.
[28] MONTES-ROMERO J, HEINZLE N, LIVERA A, et al.Novel data-driven health-state architecture for photo- voltaic system failure diagnosis[J]. Solar Energy, 2024, 279: 112820.
[29] MANSOURI M M, HADJERI S, BRAHAMI M.New method of detection, identification, and elimination of photovoltaic system faults in real time based on the adaptive neuro-fuzzy system[J]. IEEE Journal of Photovoltaics, 2021, 11(3): 797-805.
[30] 彭勃, 李耀东, 龚贤夫. 基于自编码的改进K-means光伏能源数据清洗方法[J]. 计算机科学, 2024, 51(增刊1): 725-729.
[31] 余洋, 陆文韬, 陈东阳, 等. 光伏波动平抑下改进K-means的电池储能动态分组控制策略[J]. 电力系统保护与控制, 2024, 52(7): 1-11.
[32] 黄劼, 汪逸帆, 林叶青, 等. 基于K均值聚类算法的谐振接地系统故障区段定位方法[J]. 电气技术, 2024, 25(3): 24-31, 37.
[33] 邰彬, 黄杨珏, 沈开程, 等. 基于指标加权K-means++算法的分布式光伏功率波动平抑控制方法[J]. 武汉大学学报(工学版), 2023, 56(11): 1413-1424.
[34] 邓祥力, 廖玥琳, 朱宏业, 等. 基于数字孪生模型电流动态时间规整差异度的变压器早期故障辨识[J]. 电力系统保护与控制, 2023, 51(12): 156-167.
[35] 申江卫, 岩川, 刘永刚, 等. 基于数据挖掘与大数据分析的电池故障诊断与异常检测[J]. 电工技术学报, 2024, 39(24): 7979-7994.
[36] 阳瑞霖, 莫凡, 金艳, 等. 基于重心平均动态时间规整算法的有载分接开关机械故障诊断[J]. 高电压技术, 2023, 49(4): 1515-1525.