永磁电机高频注入驱动系统逆变器非线性分析与补偿

发布时间：2017-12-27 05:05

本文关键词：永磁电机高频注入驱动系统逆变器非线性分析与补偿　出处：《哈尔滨工业大学》2015年硕士论文　论文类型：学位论文

【摘要】：永磁同步电机因具备高效率、高功率密度及良好动态性能等优点而广泛应用于工业驱动、交通运输、航空航天等领域。无位置传感器技术已成为当今电机驱动方向的研究热点,由于矢量控制依赖于电机参数的准确性,因此永磁电机离线参数辨识对于无传感器矢量控制器和观测器设计十分重要。然而,电机参数离线辨识受逆变器非线性的影响,会使得高频电流响应中含有明显的谐波误差,从而造成参数辨识误差。此外,在基于高频方波注入的无传感器控制系统中,逆变器非线性会导致转子位置估计结果中出现明显的脉动误差,进而影响电机解耦的准确性。为进一步提升永磁同步电机驱动控制系统性能,本文主要对高频信号注入的逆变器非线性分析及补偿策略进行研究。首先分析逆变器非线性对于响应电流造成的谐波误差,并研究高频加直流复合信号注入方法抑制逆变器非线性的影响。通过分析考虑逆变器非线性影响下的电机数学模型,研究基于复合信号注入的参数离线辨识策略,设计基于离散傅里叶变换的高频信号幅值提取方法,从而根据注入电压及响应电流幅值计算出电感参数。在此基础上,针对不同注入信号条件下逆变器非线性影响引起的电流响应中的2次和3次谐波分布情况进行分析,并验证其对电感辨识精度的影响。为进一步减小参数辨识误差,在对逆变器非线性特性进行自学习的基础上,提出针对不同相电流条件下逆变器非线性自适应补偿方法,从而抑制其对参数辨识的负面影响。在基于高频方波注入的永磁同步电机无位置传感器控制基础上,研究通过分离高频电流包络线提取位置特征信息的信号处理方法,并分析逆变器非线性对转子位置观测器造成的脉动误差。针对基频和高频分量综合作用引起的负面影响进行研究,通过阐明高频注入法的零电流钳位效应基础上,采用改进高频信号注入方式以及信号提取方法来抑制逆变器非线性对无传感器控制的影响。在理论分析的基础上,通过Matlab/Simulink仿真工具对所研究的参数离线辨识方法和基于方波注入的无传感器控制策略进行仿真分析,并在2.2k W内置式永磁同步电机对拖加载实验平台上进行了试验验证,结果表明了方法的有效性。
[Abstract]:Permanent magnet synchronous motor (PMSM) is widely used in industrial drive, transportation, aerospace and other fields due to its advantages of high efficiency, high power density and good dynamic performance. Sensorless technology has become the research focus of motor drive nowadays. Because vector control depends on the accuracy of motor parameters, the off-line parameter identification of permanent magnet motor is very important for sensorless vector controller and observer design. However, the off-line identification of motor parameters is influenced by the nonlinearity of inverter, which will cause obvious harmonic error in high-frequency current response, resulting in parameter identification error. Besides, in the sensorless control system based on high frequency square wave injection, the nonlinearity of the inverter will lead to the obvious pulsation error in the rotor position estimation result, which will further affect the accuracy of the motor decoupling. In order to further improve the performance of PMSM drive control system, the nonlinear analysis and compensation strategy of high frequency signal injection inverter is mainly studied in this paper. First, the harmonic error caused by the nonlinearity of the inverter is analyzed, and the influence of high-frequency and DC composite signal injection method on the nonlinearity of the inverter is studied. By analyzing the mathematical model of inverter motor considering nonlinear under the influence of the research on off-line parameter identification strategy of composite signal injection based on the extraction method of high frequency signal amplitude based on discrete Fourier transform, which according to the injected voltage and current response amplitude to calculate electric inductance. On this basis, the distribution of 2 and 3 harmonic currents in the current response caused by the nonlinear influence of the inverter under different injection signal conditions is analyzed, and the influence on the inductance identification accuracy is verified. In order to further reduce the parameter identification error, based on the self learning of the inverter's nonlinear characteristics, a nonlinear adaptive compensation method for inverter under different phase currents is put forward, so as to suppress its negative influence on parameter identification. Based on the sensorless control of PMSM Based on high frequency square wave injection, a signal processing method is presented to extract position feature information by separating the high frequency current envelope, and the pulsation error caused by the nonlinearity of the inverter to the rotor position observer is analyzed. Research on negative implications for the comprehensive effect of the fundamental frequency and the high frequency component caused by the zero current clamp of high frequency injection method based on the improved effect, high frequency signal injection method and signal extraction method to suppress the influence of nonlinearity on the inverter sensorless control. On the basis of theoretical analysis, through the Matlab/Simulink simulation tool for off-line parameter identification method based on the research and sensorless control strategy simulation analysis of square wave injection, and 2.2k W in the permanent magnet synchronous motor to drag loading on the experimental platform of experiment, the results show that the effectiveness of the method.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM351;TM464

【参考文献】