永磁同步电机转动惯量辨识方法研究

发布时间：2018-12-15 12:57

【摘要】：永磁同步电机由于高功率密度、体积小、高可靠性的特点,被广泛应用于高性能控制领域中,如工业机器人、CNC数控机床、混合动力汽车等。其中转动惯量是永磁同步电机中一个十分重要的参数。一般情况下,若负载变化,转动惯量也会变化。当电机转动惯量变化较大时,若伺服系统控制器参数不变,则会导致伺服电机的控制精度变低,甚至会引起控制系统不稳定。因此,需要辨识永磁同步电机伺服系统转动惯量,并且根据辨识的结果对电机控制器的PI参数进行整定。这样才能提高伺服系统的控制性能,并增强控制系统的抗干扰能力和鲁棒性。由于现有的辨识算法,如减速法、人工轨迹法等,都是属于离线辨识。虽然其算法简单,但是其辨识精度低,以及辨识时间长。而现有的在线辨识算法,如最小二乘法、模型参考自适应算法等,虽然有一些学者在进行研究,但是仍然不能得出令人满意的结果。因此在线辨识仍然处于研究阶段,值得深入研究。本文首先介绍了永磁同步电机的数学模型和矢量控制基本原理;其次深入研究了积分法、改进的最小二乘法、模型参考自适应辨识算法原理,并在Matlab中搭建了电机控制系统仿真模型,并采用上述不同的算法在该模型基础上进行仿真,并得出各个算法的辨识结果。再次针对不同的辨识算法结果进行分析,并根据辨识结果对各种算法进行相应地改进,再次仿真并与之前的结果进行比较,得出结论。针对传统最小二乘法辨识产生的数据饱和的现象,引入遗忘因子。仿真结果表明,利用遗忘因子最小二乘法能够有效防止数据饱和并能够使辨识时间缩短。同时针对传统模型参考自适应原理辨识误差较大的现象,引入了加权递推平均滤波算法。仿真结果表明,引入加权平均滤波算法后,不仅可以减小辨识的误差,同时也能让辨识误差和辨识时间有一个较好的折衷选择。最后根据电机控制系统的要求,选取最适合的算法进行实验的验证。并在永磁同步电机伺服控制系统中加入参数自整定功能。仿真和实验结果表明,模型参考自适应算法辨识效果较好,且基于转动惯量的控制器PI参数自整定可以显著地提高控制系统的抗干扰能力。
[Abstract]:Permanent magnet synchronous motor (PMSM) is widely used in high performance control fields such as industrial robot, CNC machine tool, hybrid electric vehicle and so on because of its high power density, small size and high reliability. The moment of inertia is a very important parameter in PMSM. In general, if the load changes, the moment of inertia will also change. When the moment of inertia of the motor changes greatly, if the parameters of the servo system controller are not changed, the control precision of the servo motor will become lower, and even the control system will be unstable. Therefore, it is necessary to identify the moment of inertia of the permanent magnet synchronous motor servo system, and to adjust the PI parameters of the motor controller according to the identification results. Only in this way can the control performance of the servo system be improved, and the anti-interference ability and robustness of the control system be enhanced. The existing identification algorithms, such as deceleration method and artificial trajectory method, are all off-line identification. Although its algorithm is simple, but its identification accuracy is low, and the identification time is long. However, the existing online identification algorithms, such as least square method, model reference adaptive algorithm and so on, have been studied by some scholars, but the results are still not satisfactory. Therefore, online identification is still in the research stage, which is worthy of further study. This paper first introduces the mathematical model and the basic principle of vector control of permanent magnet synchronous motor (PMSM). Secondly, the integration method, the improved least square method, the principle of model reference adaptive identification algorithm are deeply studied, and the simulation model of motor control system is built in Matlab, and the simulation is carried out on the basis of the above different algorithms. The identification results of each algorithm are obtained. The results of different identification algorithms are analyzed again, and the corresponding improvements are made according to the identification results, and then the simulation results are compared with the previous results, and the conclusions are drawn. Aiming at the phenomenon of data saturation caused by traditional least square method, the forgetting factor is introduced. The simulation results show that the least square method of forgetting factor can effectively prevent data saturation and shorten the identification time. At the same time, a weighted recursive average filter algorithm is introduced to solve the problem of large identification error in the traditional model reference adaptive principle. The simulation results show that the weighted average filtering algorithm can not only reduce the identification error, but also make a good compromise between the identification error and the identification time. Finally, according to the requirements of motor control system, select the most suitable algorithm for experimental verification. The parameter self-tuning function is added to the servo control system of permanent magnet synchronous motor (PMSM). The simulation and experimental results show that the model reference adaptive algorithm is effective and the PI parameter self-tuning based on the moment of inertia can significantly improve the anti-jamming ability of the control system.
【学位授予单位】：东华大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TM341

【相似文献】