基于自适应高斯基表示的神经网络在电力系统故障和振荡识别中的应用Application of Neural Network Based on Adaptive Gaussian Representation in Discrimination of Fault and Oscillation in Power System
熊卫华;赵光宙;
摘要(Abstract)：
结合最优联合时一频处理无交叉项干扰及神经网络自学习分类识别的优点,提出了一种在有色噪声干扰下识别电力系统故障和振荡的方法。将经过自适应高斯基表示(Adaptive Gaussian Representation,AGR)分析处理的电力信号特征向量输入神经网络分类器进行识别。待辨识输入向量不仅表征了原信号的基本信息,而且没有交叉项,运算简单。仿真结果表明,此方法能正确分类识别有色噪声干扰下的系统故障和振荡,提高了电力系统微机保护在系统振荡中检测故障的灵敏性和精确性。
关键词(KeyWords)： 自适应高斯基表示;神经网络;电力系统;故障;振荡
基金项目(Foundation):
作者(Author): 熊卫华;赵光宙;
Email:
DOI: 10.13335/j.1000-3673.pst.2006.03.005
参考文献(References)：
- [1] 孔繁鹏，葛耀中，周秦武．一种区分振荡与故障的新方法[J]．电力系统自动化，1995，19(4)：34-38．Kong Fanpeng, Ge Yaozhong, Zhou Qinwu. A new principle of discrimination between power swing and short circuit[J]. Automation of Electric Power Systems, 1995, 19(4): 34-38.
- [2] 林湘宁，刘沛．全电流与故障分量电流比例差动判据的比较研究[J]．中国电机工程学报，2004，24(10)：27-31．Lin Xiangning, Liu Pei. Comparative studies on percentage differential criteria using phase current and superimposed phase current[J]. Proceedings of the CSEE, 2004, 24(10): 27-31.
- [3] 李晓华，张哲，尹项根，等．故障分量比率差动保护整定值的选取[J]．电网技术，2001，25(4)：47-50．
- Li Xiaohua, Zhang Zhe, Yin Xianggen, et al. Selection of settings of differential protection based on fault component[J]. Power System Technology, 2001, 25(4): 47-50.
- [4] Scholtz E, Sonthikorn P, Verghese G C. Observers for dynamic state estimation of swing motions in power systems[C]. Power Systems Computation Conference, Sevilla, 2002.
- [5] 许庆强，索南加乐，宋国兵，等．振荡时电力系统瞬时频率的实时测量[J]．中国电机工程学报，2003，23(1)：51-55．Xu Qingqiang, Suonan Jiale, Song Guobin, et al. Real time measurement of power system instantaneous frequency while power system swings[J]. Proceedings of the CSEE, 2003, 23(1): 51-55.
- [6] 夏明超，黄益庄，王勋，等．高压输电线路暂态保护的发展和现状[J]．电网技术，2002，26(11)：65-69．Xia Mingchao, Huang Yizhuang. Wang Xun, et al. Development and present situation of transient based protections for high voltage power transmission lines[J]. Power System Technology, 2002, 26(11): 65-69.
- [7] 鲍小鹏，张举．应用模糊集理论识别电力系统振荡中不对称故障的新方法[J]．电网技术，2004，28(12)：25-29．Bao Xiaopeng, Zhang Ju. A new approach to identify unsymmetrical fault occurred during power system oscillation by use of fuzzy set theory[J]. Power System Technology, 2004, 28(12): 25-29.
- [8] 范春菊，张兆宁，郁惟镛，等．模拟神经网络在区分电力系统故障和振荡中的应用[J]．电力系统及其自动化学报，2001，13(3)：50-53．Fan Chunju, Zhang Zhaoning, Yu Weiyong, et al. The application of the fuzzy neural network in the distinguishing of the fault and the oscillation[J]. Proceedings of the EPSA, 2001, 13(3): 50-53.
- [9] 毛鹏，张兆宁，林湘宁，等．基于小波神经网络的电力系统振荡和故障识别[J]．电力系统自动化，2002，26(11)：9-13．Mao Peng, Zhang Zhaoning, Lin Xiangning, et al. Wavelet neural networks based recognition of swing and fault in power system [J]. Automation of Electric Power Systems, 2002, 26(11): 9-13.
- [10] 欧阳森，宋政湘，陈德桂，等．小波软阈值去噪技术在电能质量检测中的应用[J]．电力系统自动化，2002，26(19)：56-60．Ouyang Sen, Song Zhengxiang, Chen Degui, et al. Application of wavelet soft-threshold de-nosing technique to power quality detection[J]. Automation of Electric Power Systems, 2002, 26(19):56-60.
- [11] 周银春，孙才新，杜林，等．用小波剔除局放信号中白噪声的一种实用方法[J]．重庆大学学报(自然科学版)，2002，25(3)：68-71．Zhou Yinchun, Sun Caixin, Du Lin, et al. A practical method of eliminating the white noise of partial discharge with wavelet [J]. Journal of Chongqing University(Natural Science Edition), 2002, 25(3): 68-71.
- [12] 张贤达．现代信号处理[M]．北京：清华大学出版社，1995．
- [13] Shie Qian, Dapang Chen. Signal representation using adaptive normalized Gaussian functions[J]. Signal Processing, 1994, 36(1):1-11.
- [14] Kim K T, Choi I S, Kim H T. Efficient radar target classification using adaptive joint T-F processing[J]. IEEE Trans on Antennas and Propagation, 2000, 48(12): 1789-1801.
- [15] Trintinalia L C, Ling H. Joint time-frequency ISAR using adaptive processing[J]. IEEE Trans on Antennas and Propagation, 1997, 45(2): 221-227.
- [16] 熊卫华，赵光宙，厉小润．基于AGR提取PD信号的应用研究[J]．电力系统及其自动化学报，2005，17(1)：31-35．Xiong Weihua, Zhao Guangzhou, Li Xiaorun. Extracting partial discharge signals using AGR[J]. Proceedings of the EPSA, 2005, 17(1): 31-35.
- [17] 孙增圻，张再兴，邓志东，等．智能控制理论及技术[M]．北京：清华大学出版社，1997．