基于人工免疫网络的断路器在线自学习故障诊断.pdfVIP

第29 卷 第34 期 中 国 电 机 工 程 学 报 Vol.29 No.34 Dec. 5, 2009 128 2009 年12 月5 日 Proceedings of the CSEE ©2009 Chin.Soc.for Elec.Eng. 0258-8013 (2009) 34-0128-07 TM 561 A 470 40 文章编号： 中图分类号： 文献标志码： 学科分类号： ⋅ 基于人工免疫网络的断路器在线自学习故障诊断 1 2 1 吕超，于洪海，王立欣 (1．哈尔滨工业大学电气工程及自动化学院，黑龙江省 哈尔滨市 150001 ； 2．哈尔滨超高压局，黑龙江省 哈尔滨市 150090) On-line Self-learning Fault Diagnosis for Circuit Breakers Based on Artificial Immune Network 1 2 1 LÜ Chao , YU Hong-hai , WANG Li-xin (1. School of Electrical Engineering Automation, Harbin Institute of Technology, Harbin 150001, Heilongjiang Province, China; 2. Harbin Ultra High Voltage Bureau, Harbin 150090, Heilongjiang Province, China) ABSTRACT: The existing diagnostic methods for high 证设备安全运行，降低寿命周期费用，具有较大的 voltage circuit breakers lack in the ability to pursue the change 现实意义。 of mechanical state. An on-line self-learning classifier, C-aiNet, 断路器机械状态监测的对象包括分/合闸时间、 for identifying mechanical failures of HVCBs based on 相内/相间不同期、刚分/ 刚合速度、动触头行程、 artificial immune network aiNet is presented. The training and 脱扣线圈电流和控制箱振动加速度等[3-8] 。其中振动 self-learning process were introduced in details, and principles 信号由断路器动作过程中的一系列暂态振动事件 for selecting network parameters were discussed as well. This method can trace the new regions of clusters, discard old ones, 构成，包含了相关机械部件的状态信息，从振动信 and recognize new patterns. The results of simulation based on 号中提取特征量，实现断路器机械故障诊断是当 measured vibration characteristic data of HVCBs show tha