基于大数据分类与SVDD模型的多场景500kV劣化绝缘子判定研究  

Multi-Scenario Judgment of 500 kV Deteriorated Insulator Based on Big Data Classification and SVDD Model

作　　者：普子恒[1,2] 彭朕 方春华 郑雷[3,4] 罗浩 殷鹏翔[3] PU Ziheng;PENG Zhen;FANG Chunhua;ZHENG Lei;LUO Hao;YIN Pengxiang(College of Electrical Engineering and New Energy,China Three Gorges University,Yichang 443002,China;Hubei Transmission Line Engineering Technology Research Center,Yichang 443002,China;Wuhan NARI Limited Company,State Grid Electric Power Research Institute,Wuhan 430074,China;Hubei Key Laboratory of Power Grid Lightning Risk Prevention,Wuhan 430074,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002 [2]湖北省输电线路工程技术研究中心,湖北宜昌443002 [3]国网电力科学研究院武汉南瑞有限责任公司,武汉430074 [4]电网雷击风险预防湖北省重点实验室,武汉430074

出　　处：《电瓷避雷器》2022年第2期189-196,共8页Insulators and Surge Arresters

基　　金：武汉市基金项目“武汉市‘3531光谷人才计划’”(编号:524625190017)。

摘　　要：正常状态下500 kV交流瓷绝缘子串在不同场景时分布电压差异较大,根据现有DL/T 487标准电压对所有场景下的绝缘子串进行劣化判定容易产生误判。随着智能运维设备的开发与应用,可为劣化绝缘子判定研究提供数据支持。为准确判定劣化绝缘子,笔者首先建立500 kV绝缘子串多场景下分布电压仿真计算模型,仿真表明塔型、相位、绝缘子片数、均压环是绝缘子串分布电压的重要影响因素;进一步基于这些因素建立大数据分类与SVDD模型的多场景下500 kV劣化绝缘子判定模型,利用仿真得到的分布电压作为样本数据,对模型进行训练和测试。结果表明:基于大数据分类与SVDD模型的多场景500 kV劣化绝缘子判定模型可以对绝缘子进行劣化判定,判定准确率可达到96%。Under normal conditions, the distribution voltage of 500 kV AC porcelain insulator strings varies greatly in different scenarios. It is easy to misjudge the deterioration of insulator strings in all scenarios according to the existing DL/T 487 standard voltage. With the development and application of intelligent operation and maintenance equipment, it can provide data support for the determination of degraded insulators. In order to accurately determine the deteriorated insulator, the author first establishes the distributed voltage simulation model of 500 kV insulator string under multiple scenarios. The simulation shows that tower type, phase, number of insulator pieces and voltage equalizing ring are important factors affecting the distributed voltage of insulator string;further, based on these factors, a 500 kV degraded insulator judgment model under multiple scenarios of big data classification and SVDD model is established. The distributed voltage obtained by simulation is used as sample data to train and test the model. The results show that the multi scenario 500 kV degraded insulator judgment model based on big data classification and SVDD model can judge the deterioration of insulators, and the judgment accuracy can reach 96%.

关 键 词：500KV交流输电线路 绝缘子检测 分布电压 多场景 大数据分类 SVDD模型 

分 类 号：TM216[一般工业技术—材料科学与工程]