基于RBFNN识别的电力系统动态参数获取新途径.pdfVIP

26 卷 10 期 中 国 电 机 工 程 学 报 Vol.26 No.10 May 2006 2006 年5 月 Proceedings of the CSEE 2006 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 (2006) 10-0104-05 中图分类号：TM73 文献标识码：A 学科分类号：47040 基于径向基函数神经网络识别的电力系统 动态参数获取新途径 赵书涛，李宝树，苑津莎，赵东生 (华北电力大学电力系统保护与动态安全监控教育部重点实验室, 河北省 保定市 071003) The New Acquiring Path of Power System Dynamic Parameter Based on RBFNN Recognition ZHAO Shu-tao, LI Bao-shu, YUAN Jin-sha, ZHAO Dong-sheng (Key Laboratory of Power System Protection and Dynamic Security Monitoring and Control under Ministry of Education, North China Electric Power University, Baoding 071003, Hebei Province, China) ABSTRACT : Acquiring the dynamic parameters based on 测量和测量方法的限制，需过多地考虑系统安全、 computer vision instead of manual patrol, and via comparing 绝缘和弱信号的传输等因素。目前安装的多种微机 power meter reading with its preliminary definition threshold, 监测系统本身也不够完善，不仅运行平台和通信规 which is a new automatic monitoring scheme of power system 约不统一，且现场安装接线复杂、维护困难，更严 running. In the process of power meter automatic recognition, 重的是通信接口一旦 坏则无法获取监测量。此外， three particle feature invariants, include length ratio, compactness and simplicity factor have been extracted. The 有些运行参数和故障征兆信号很难通过接触测量 [1-2] RBFNN is utilized in the dial plate elements recognition. The 转换成电信号, 甚至无法利用微机监测获取 。 meter identifying experiments proved that the RBFNN input 为此，人工定时巡视中需观测多个运行参数， invariants are stable and proper, and it is insensitive to the 这作为判