六相永磁同步电机控制器故障诊断

发布时间：2018-03-17 08:43

本文选题：六相永磁同步电机　切入点：故障诊断　出处：《沈阳工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：由于六相永磁同步电机具有低压大功率、转矩脉动小、系统可靠性高等优点,因此被广泛应用于全电力舰船推动系统、纯电动以及混合动力电动车辆牵引系统等高功率等级、高可靠性的场合。六相永磁同步电机控制系统一般由电机、控制器和传感器等组成,其中控制器是容易出现故障的环节,其可靠性对整个系统的正常运行非常重要。因此,对六相永磁同步电机控制器进行早期故障诊断就显得非常重要。本文给出了基于小波包分析、流形学习法和支持向量机相结合的“特征提取→维数约简→模式识别”的六相永磁同步电机控制器故障诊断方法。首先,建立六相永磁同步电机数学模型,并仿真实现其矢量控制。介绍六相永磁同步电机的结构及其工作原理,推导出自然坐标系下和两相旋转坐标系下的数学模型,并且在Simulink中仿真实现六相永磁同步电机数学模型及其矢量控制。其次,采用小波包分析提取电机六相定子电流中的故障特征向量。详细地介绍小波分析理论和小波包分析理论。采样不同IGBT故障状态下的六相定子电流,采用小波包分析对电流进行六层小波包分解,计算各个节点的能量值,并将该能量值归一化处理后作为故障特征向量。再次,通过流形学习方法对六相永磁同步电机控制器故障特征向量进行维数约简。详细的介绍流形学习的相关理论及其几种经典算法,并采用流形学习法中局部切空间排列算法对小波包分析提取的特征向量进行降维处理。最后,利用最小二乘支持向量机对六相永磁同步电机控制器IGBT故障进行识别。介绍支持向量机理论,以及在支持向量机理论上进一步发展的最小二乘支持向量机理论。局部切空间算法约简后的特征向量分别作为最小二乘支持向量机的训练样本和测试样本,通过仿真实验对不同的IGBT开路故障进行验证,仿真验证表明,这种故障诊断的方法在理论上可以实现对不同IGBT开路故障的准确定位。
[Abstract]:Due to its advantages of low voltage and high power, low torque ripple and high reliability, six-phase permanent magnet synchronous motor is widely used in high power class, such as full electric ship propulsion system, pure electric and hybrid electric vehicle traction system, etc. The control system of six-phase permanent magnet synchronous motor is generally composed of motor, controller and sensor, among which the controller is prone to failure, and its reliability is very important for the normal operation of the whole system. It is very important to diagnose the fault of six-phase PMSM controller in the early stage. In this paper, the feature extraction based on wavelet packet analysis, manifold learning and support vector machine is presented. 鈫扗imension reduction. 鈫扵he fault diagnosis method of six-phase permanent magnet synchronous motor controller based on pattern recognition is introduced. Firstly, the mathematical model of six-phase permanent magnet synchronous motor is established and its vector control is simulated. The structure and working principle of six-phase permanent magnet synchronous motor are introduced. The mathematical models in natural coordinate system and two-phase rotating coordinate system are derived, and the mathematical model and vector control of six-phase permanent magnet synchronous motor are simulated in Simulink. The wavelet packet analysis is used to extract the fault eigenvector from the six-phase stator current of the motor. The wavelet analysis theory and wavelet packet analysis theory are introduced in detail. The six-phase stator current in different IGBT fault states is sampled. The wavelet packet analysis is used to decompose the current into six layers of wavelet packet, calculate the energy value of each node, and normalize the energy value as the fault eigenvector. Through manifold learning method, dimension reduction of fault eigenvector of six-phase PMSM controller is carried out. The related theory of manifold learning and several classical algorithms are introduced in detail. And the local tangent space arrangement algorithm in manifold learning method is used to reduce the dimension of the feature vector extracted by wavelet packet analysis. Finally, Using least square support vector machine (LS-SVM) to identify the IGBT fault of six-phase PMSM controller, the theory of support vector machine (SVM) is introduced. And the least squares support vector machine (LS-SVM) theory, which is further developed in support vector machine (SVM) theory, is used as the training sample and test sample of LS-SVM, respectively, which is reduced by local tangent space algorithm. Different open circuit faults of IGBT are verified by simulation experiments. The simulation results show that the method of fault diagnosis can accurately locate the open circuit faults of different IGBT in theory.
【学位授予单位】：沈阳工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM341;TP18

【参考文献】