基于特征挖掘的电网故障诊断方法.pdfVIP

第 30 卷 第 10 期 中 国 电 机 工 程 学 报 Vol.30 No.10 Apr. 5, 2010 16 2010 年 4 月 5 日 Proceedings of the CSEE ©2010 Chin.Soc.for Elec.Eng. 0258-8013 (2010) 10-0016-07 TM 72 A 470 40 文章编号： 中图分类号： 文献标志码： 学科分类号： ⋅ 基于特征挖掘的电网故障诊断方法 李再华，白晓民，周子冠，许婧，李晓珺，张霖，孟珺遐，朱宁辉 ( 中国电力科学研究院，北京市 海淀区 100192) Method of Power Grid Fault Diagnosis Based on Feature Mining LI Zai-hua, BAI Xiao-min, ZHOU Zi-guan, XU Jing, LI Xiao-jun, ZHANG Lin, MENG Jun-xia, ZHU Ning-hui (China Electric Power Research Institute, Haidian District, Beijing 100192, China) ABSTRACT: The two main bottlenecks in the application of 的动作以及电压、电流等信息进行全局分析，确定 expert system are: the maintenance of rule base; the 故障位置和故障性质，判断保护与开关动作的正确 coordination of speed and accuracy of reasoning. Features and 性，并辅助调度员进行决策，以尽快控制故障，尽 key events of fault events were analyzed, then a novel method 量避免故障发展。 of association rule mining based on feature mining was 人工智能技术在故障诊断中应用最广[1-2]，其中 presented, the method was originated from frequent pattern (FP)-algorithm and was improved. The improvements include: 基于规则推理的专家系统应用相对最多，其主要问 features of fault information are utilized, such as the time 题是规则库难以适应电网的变化、推理的速度和准 sequence and causality of events, fault type and serious fault or 确度难以同时令人满意等。如何高效地利用日益增 unusual fault; OR logical function of rules is added; prune 加的监控信息来完善规则库，并提高推理的速度和 technique of FP-tree is improved. Use case shows the improved 准确度，是专家系统实用化面临的关键问题[2] 。 algorithm could reduce invalid mining largely, and the speed 电网在不断发展，运行方式也经常变化，因此 and acc