基于递推最小二乘法的永磁伺服系统参数辨识

发布时间：2019-03-29 14:18

【摘要】：为使永磁同步电机(PMSM)控制系统在复杂环境中具有较好的动态性能,伺服系统必须具有参数辨识和参数自整定的功能,而转动惯量与负载转矩辨识是其首要解决的问题。采用零阶保持器对电机运动方程进行离散化建模,考虑了摩擦系数对辨识结果的影响,将基于遗忘因子递推最小二乘辨识算法应用于该离散模型可以同时辨识出系统转动惯量、负载转矩和摩擦系数。同时,针对Matlab/Simulink中库模型参数不能在线动态修改的缺点,提出改进型PMSM模型,以此搭建了伺服系统的仿真控制模型,完成了定参数与变参数的动态仿真。最后,在stm32微控制器上进行了实验验证。仿真和实验表明该文提出的电机离散化模型和参数辨识方法具有一定的准确性和实时性,仿真结果验证了改进型PMSM模型在变参数仿真研究中的实用性。
[Abstract]:In order to make the permanent magnet synchronous motor (PMSM) (PMSM) control system have better dynamic performance in complex environment, servo system must have the function of parameter identification and parameter self-tuning, and the identification of moment of inertia and load torque is the first problem to be solved. The motion equation of motor is discretized by zero-order holding device, and the influence of friction coefficient on the identification result is considered. The recursive least square identification algorithm based on forgetting factor is applied to the discrete model to identify the moment of inertia of the system at the same time. Load torque and friction coefficient. At the same time, aiming at the disadvantage that the parameters of library model can not be dynamically modified on-line in Matlab/Simulink, an improved PMSM model is proposed to build the simulation control model of servo system, and the dynamic simulation of fixed parameters and variable parameters is completed. Finally, the experiment is carried out on stm32 microcontroller Simulation and experiments show that the discrete model and parameter identification method proposed in this paper have certain accuracy and real-time performance. The simulation results verify the practicability of the improved PMSM model in the simulation of variable parameters.
【作者单位】：湖州师范学院工学院;南京航空航天大学江苏省新能源发电与电能变换重点实验室;
【基金】：国家自然科学基金(61573136,61573137) 湖州市公益性技术应用研究计划项目(2015GZ05)资助
【分类号】：TM341;TM921.541

【相似文献】