【作者】 许伯强；

【导师】 李和明；

【作者基本信息】 华北电力大学（河北） ， 电力系统及其自动化， 2003， 博士

【摘要】 本文对异步电动机转子断条与定子绕组匝间短路故障特征进行了机理、仿真与实验分析。在此基础上，提出了异步电动机转子断条与定子绕组匝间短路故障检测新方法；首次提出了笼型异步电动机转子断条与定子绕组匝间短路故障联合检测方法；基于参数辨识技术，提出了异步电动机定转子绕组过热故障检测新方法。仿真与实验结果表明，以上方法是切实可行的。本文的主要研究成果如下：1. 详细推导了YN接、Y接、D接笼型异步电动机在正常、转子断条及定子绕组匝间短路故障情况下的多回路数学模型。2. 对笼型异步电动机转子断条故障特征做了机理分析，以笼型异步电动机多回路数学模型为基础，首次完成了转子导条持续断裂故障的暂态仿真，同时对转子断条故障进行了实验研究。通过对大量仿真、实验数据的详尽分析，系统而全面地总结了转子断条故障特征。在此基础上，提出了基于连续细化傅里叶变换、神经网络与自适应滤波技术的笼型异步电动机转子断条故障检测新方法。3. 对异步电动机定子绕组匝间短路故障特征进行了机理、仿真与实验分析。基于神经网络技术，提出了以定子负序导纳平均值作为故障特征量的异步电动机定子绕组匝间短路故障检测方法。基于连续细化傅里叶变换技术,提出了以定子绕组(等效)阻抗角作为故障特征量的异步电动机定子绕组匝间短路故障检测新方法，首次指出若定子某相绕组发生匝间短路故障，则其相邻滞后相（等效）阻抗角在定子三相绕组中恒为最大，据此可以识别定子绕组匝间短路故障发生相。4. 首次提出了笼型异步电动机转子断条与定子绕组匝间短路故障联合检测方法，应用该方法可以正确识别笼型异步电动机转子断条与定子绕组匝间短路故障，进一步提高故障检测可靠性。5. 基于异步电动机坐标系统派克数学模型，推导出异步电动机对称稳态运行时的定转子电阻辨识算法。在此基础上，提出了基于参数辨识技术的异步电动机定转子绕组过热故障检测新方法。6. 研制成功异步电动机绕组故障检测系统，于2001年10月在山西娘子关发电厂投入运行并获得好评。该系统已通过现场验收。更多还原

【Abstract】 The winding faults, such as rotor bar breaking, stator winding inter-turn short circuit and stator/rotor over-heating, are prevalent and potentially destructive in asynchronous motors. This paper focuses on the early detection of the winding faults in asynchronous motors, as would help avoid catastrophic failure, reduce repair cost and outage time to a great extent. Firstly, the fault features of rotor bar breaking and stator winding inter-turn short circuit are analyzed in this paper, deeply and thoroughly. Secondly, a series of novel detection schemes are presented creatively. Furthermore, this paper proposes the joint detection scheme for rotor bar breaking and stator winding inter-turn short circuit fault in squirrel cage asynchronous motors, for the first time. Meanwhile, a parameter identification technique based approach to detect stator/rotor over-heating fault is also presented. The validity and feasibility of the above-mentioned detection schemes has been verified by a great deal simulation and experiment results. Finally, the detection system for the winding faults in asynchronous motors has been developed successfully. The distinguished contributions of this paper are listed in details as the following.1. The multi-loop mathematical models under the conditions that the squirrel cage asynchronous motor is healthy, faulty with broken rotor bars or with stator winding inter-turn shorts are deduced thoroughly, and improved necessarily. And thus, the numerical simulation method, which is valid and practical, is formed. 2. For the first time, the transient simulation of rotor bar subsequent-breaking fault has been completed successfully. Also, the corresponding experiments have been carried out. Moreover, this paper systematically generalizes the rotor bar breaking fault features, and draws quite a few creative and interesting conclusions. Based on continuous subdivision Fourier transform, artificial neural network and self-adaptive filter technique, this paper presents a novel detection scheme for rotor bar breaking fault. Results of simulation and experiments demonstrate that the detection scheme is valid and feasible.The transient simulation of stator winding inter-turn short circuit fault has been completed successfully based on the multi-loop mathematical model. Also, the corresponding experiments have been carried out. According to that, this paper generalizes the stator winding inter-turn short circuit fault features systematically. By treating the average of stator negative sequence admittance as the fault feature, this paper proposes an artificial neural network based approach to detect stator winding inter-turn short circuit fault, whose sensitivity and reliability are satisfactory. At the same time, a continuous subdivision Fourier transform based approach to detect stator<WP=5>3. winding inter-turn short circuit fault, which selects the stator winding (equivalent) impedance angle as the fault feature, is presented. Furthermore, this paper points out that the lagging stator winding, adjacent to the faulty stator winding with inter-turn shorts, possesses the largest impedance angle among the three-phase stator windings, as allows to identify the faulty stator winding.4. The joint detection necessity for rotor bar breaking and stator winding inter-turn short circuit fault in squirrel cage asynchronous motors is clarified, and the corresponding joint detection strategy, valid and practical, is recommended for the first time, in this paper.5. A novel stator/rotor over-heating detection approach based on parameter identification technique is presented in this paper. Firstly, the stator/rotor resistances are identified on-line. Secondly, the corresponding temperatures are figured out according to the principle that the metal resistance linearly depends on its temperature. Finally, stator/rotor over-heating detection is accomplished.6. The detection system for the winding faults in asynchronous motors has been developed successfully, whose on-site application in Niang Ziguan power更多还原