王攀攀,史丽萍,杨晓冬,张涛

• 1:中国矿业大学信息与电气工程学院

摘要(Abstract)：

为了准确识别感应电机定子匝间短路故障,该文提出一种基于骨干微粒群算法优化支持向量机的故障诊断方法,并给出了可行的诊断步骤。该方法首先利用小波包频带能量分解技术,将定子电流信号的各频率分量分解到不同频带,形成感应电机运行状态的特征向量,并以此作为支持向量机的输入向量。采用支持向量机进行分类,并利用无需设置控制参数的骨干微粒群算法和交叉检验优化模型参数,避免了参数选择的盲目性。最后试验结果表明,该方法诊断感应电机定子匝间短路故障能取得良好的效果。

关键词(KeyWords)： 感应电机;定子匝间短路;骨干微粒群优化算法;小波包;支持向量机;故障诊断

Abstract:

Keywords:

基金项目(Foundation): 教育部科学技术研究重大项目(311021)

作者(Author): 王攀攀,史丽萍,杨晓冬,张涛

参考文献(References)：

• [1]Nandi S,Toliyat H A.Condition monitoring and fault diag-nosis of electrical machines-a review[C]//IEEE Industry Applications Conference.Phoenix,USA:1999.
• [2]宋功益,郭清滔,涂福荣,等(Song Gongyi,Guo Qingtao,Tu Furong,et al).模糊贝叶斯网的变压器故障诊断(Novel method for transformer faults diagnosis based on theory of fuzzy Bayesian networks)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2012,24(2):102-106.
• [3]曲正伟,荣亚军,刘帅,等(Qu Zhengwei,Rong Yajun,LiuShuai,et al).RBF-NN对发电机转子绕组匝间短路的诊断(RBF-NN’s diagnosis of generator rotor winding inter-turn short circuit fault)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2011,23(1):114-117.
• [4]吕干云,程浩忠,董立新,等(LüGanyun,Chen Haozhong,Dong Lixin,et al).基于多级支持向量机分类器的电力变压器故障识别(Fault diagnosis of power transformer based on multi-layer SVM classifier)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2005,17(1):19-22,52.
• [5]朱苏航,吕干云(Zhu Suhang,LüGanyun).利用遗传支持向量机进行电压暂降信号识别(Voltage sag signal i-dentification with GA-SVM)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2012,24(1):84-87.
• [6]Kennedy J.Bare bones particle swarms[C]//IEEE Swarm Intelligence Symposium.Indianapolis,USA:2003.
• [7]Kennedy J,Eberhart R.Particle swarm optimization[C]//International Conference on Neural Networks Proceedings.Perth,USA:1995.
• [8]Clerc M,Kennedy J.The particle swarm-explosion,stabil-ity,and convergence in a multidimensional complex space[J].IEEE Trans on Evolutionary Computation,2002,6(1):58-73.
• [9]Cortes C,Vapnik V.Support-vector networks[J].MachineLearning,1995,20(3):273-297.
• [10]杨俊燕,张优云,赵荣珍(Yang Junyan,Zhang Youyun,Zhao Rongzhen).支持向量机在机械设备振动信号趋势预测中的应用(Application of support vector machines in trend prediction of vibration signal of mechanical equip-ment)[J].西安交通大学学报(Journal of Xi’an Jiaotong University),2005,39(9):950-953.
• [11]Stavrou A,Sedding H G,Penman J.Current monitoring for detecting inter-turn short circuits in induction motors[J].IEEE Trans on Energy Conversion,2001,16(1):32-37.
• [12]Gentile G,Meo S,Ometto A.Induction motor current sig-nature analysis to diagnostics of stator short circuits[C]//IEEE International Symposium on Diagnostics for Electric Machines,Power Electronics and Drives.Atlanta,USA:2003.