| 引用本文: | 党晓强,邰能灵,王海田,黄 彬.基于机端测量阻抗特征和Elman动态神经网络的发电机转子绕组短路在线识别[J].电力自动化设备,2012,32(6): |
| DANG Xiaoqiang,TAI Nengling,WANG Haitian,HUANG Bin.NN-based online identification of rotor inter-turn short circuit faults according to generator terminal impedance[J].Electric Power Automation Equipment,2012,32(6): |
|
| 摘要: |
| 转子绕组短路会在一定程度上引起主极磁势缺失。提出将发电机失磁故障初始阶段中的等有功阻抗圆概念与转子绕组短路联系，针对失磁故障初始阶段机端测量阻抗的变化建立故障特征样本，应用Elman动态神经网络对在线故障信号进行识别诊断。在MATLAB6.5软件下对一台大型两极同步发电机转子进行绕组短路的仿真得到故障数据，将机端阻抗幅值和等效机端阻抗相角的无功功率作为故障特征输入Elman网络进行识别，仿真结果表明该技术可以在线识别出发电机转子短接匝数的百分比，判断短路的严重程度。 |
| 关键词: 转子 短路 故障分析 发电机 阻抗 神经网络 诊断 |
| DOI： |
| 分类号: |
| 基金项目:国家教育部留学归国人员科研启动基金资助项目（教外司留［2011］1139号）；省部共建教育部重点实验室学术成果培育项目（SBZDPY-11-15） |
|
| NN-based online identification of rotor inter-turn short circuit faults according to generator terminal impedance |
|
DANG Xiaoqiang, TAI Nengling, WANG Haitian, HUANG Bin
|
|
School of Electronic Information and Electrical Engineering，Shanghai Jiao Tong University，Shanghai 200240，China
|
| Abstract: |
| The inter-turn short circuit of generator rotor will cause the loss of winding magnet potential to a certain extent. It is proposed to set the fault characteristic sample based on the variation of terminal impedance during the initial phase of magnet loss to build the relationship between the impedance circle of equal active power and the inter-turn short circuit faults of rotor windings，based on which，Elman NN（Neural Network） is applied to online identify the faults. A large salient synchronous generator with rotor inter-turn short circuit is simulated with MATLAB6.5 and the amplitude of terminal impedance and the inactive power of equivalent phase are collected as the characteristic data of faults，which are used to train Elman NN. Simulative results indicate that，the percentage of inter-turn short circuit can be online detected and the seriousness of rotor winding fault is estimated. |
| Key words: short circuit failure analysis electric generators electric impedance neural networks diagnosis |
