中国电力 ›› 2025, Vol. 58 ›› Issue (6): 76-82.DOI: 10.11930/j.issn.1004-9649.202408049

• 基于数据驱动的电力系统安全稳定分析与控制 • 上一篇    下一篇

面向智能变电站保护的FMM测试及改进SVM校验诊断技术

吕鹏飞1(), 裘愉涛2(), 金盛3(), 王志华4()   

  1. 1. 国家电网有限公司,北京 100032
    2. 国网浙江省电力有限公司,浙江 杭州 310007
    3. 国网嘉兴供电公司,浙江 嘉兴 314000
    4. 武汉凯默电气有限公司,湖北 武汉 430223
  • 收稿日期:2024-08-14 发布日期:2025-06-30 出版日期:2025-06-28
  • 作者简介:
    吕鹏飞(1979),男,硕士,高级工程师(教授级),从事交直流系统继电保护研究,E-mail:lv-pengfei@sgcc.com
    裘愉涛(1967),男,硕士,高级工程师(教授级),从事继电保护原理及应用研究,E-mail:zdqyt@vip.sina.com
    金盛(1990),男,学士,高级工程师,从事继电保护原理及应用研究,E-mail:jsxsy0912@163.com
    王志华(1972),男,通信作者,博士,高级工程师(教授级),从事变电站智能测试及运维技术研究,E-mail:1259107825@qq.com
  • 基金资助:
    国家电网有限公司总部科技项目资助(5108-202218280A-2-268-XG)。

FMM Testing and Improved SVM Diagnostic Verification Technology for Intelligent Substation Protection

LV Pengfei1(), QIU Yutao2(), JIN Sheng3(), WANG Zhihua4()   

  1. 1. State Grid Corporation of China, Beijing 100032, China
    2. State Grid Zhejiang Electric Power Co., Ltd., Hangzhou 310007, China
    3. State Grid Jiaxing Power Supply Company, Jiaxing 314000, China
    4. Wuhan Kemov Electric Co., Ltd., Wuhan 430223, China
  • Received:2024-08-14 Online:2025-06-30 Published:2025-06-28
  • Supported by:
    This work is supported by the Science and Technology Project of State Grid Corporation (No.5108-202218280A-2-268-XG).

摘要:

为解决智能变电站大量二次设备测试复杂、工期长以及缺乏测试结果诊断的难题,引入正向最大匹配(forward maximum matching,FMM)算法实现保护自动测试,并基于蛙跳算法(frog leaping algorithm,FLA)改进支持向量机(support vector machines,SVM)模型完成测试结果的校验诊断。面向智能站保护配置信息关系,提出了保护自动测试与诊断系统架构。通过分析不同保护原理应用电压/电流定值关系建立保护测试需求,构建测试模板集合,利用哈希和编辑距离算法进行相似度比较,基于FMM算法实现了待测装置自动测试。算例分析与工程实例验证了所提技术的有效性。

关键词: 智能变电站, 保护自动测试技术, 正向最大匹配算法

Abstract:

To address the problems of complex testing, long construction period, and insufficient diagnostic testing results for numerous secondary devices in smart substations, this paper introduces the forward maximum matching (FMM) algorithm to achieve automatic protection testing, and improves the support vector machines (SVM) model based on the frog leaping algorithm (FLA) to complete the verification and diagnosis of test results. For the protection configuration and information relationship of intelligent stations, propose an architecture for protecting automatic testing and diagnosis systems; By analyzing different protection principles and the relationship between applied voltage/current and setpoints, establish protection testing requirements, construct a set of testing templates, compare similarity using hash and edit distance algorithms, and FMM algorithm was used to achieve automatic testing of the device under test. The effectiveness of the proposed technology was verified through case analysis and engineering examples.

Key words: intelligent substation, protection automatic testing technology, forward maximum matching


AI


AI小编
您好!我是《中国电力》AI小编,有什么可以帮您的吗?