基于FNN的车用永磁同步电机转动惯量识别与摩擦补偿控制  

FNN-based moment of inertia identification and friction compensation control for PMSM

作　　者：刘晏 李刚[1] 俞兆起 程浩宁 LIU Yan;LI Gang;YU Zhaoqi;CHENG Haoning(School of Automobile and Traffic Engineering,Liaoning University of Technology,Jinzhou 121000,China)

机构地区：[1]辽宁工业大学汽车与交通工程学院,辽宁锦州121000

出　　处：《重庆理工大学学报(自然科学)》2025年第2期47-54,共8页Journal of Chongqing University of Technology:Natural Science

基　　金：国家自然科学基金面上项目(51675257);辽宁省自然科学基金面上项目(2022-MS-376)。

摘　　要：考虑到伺服系统在不确定性摩擦影响下难以达到期望的控制效果,摩擦现象给电机参数辨识带来不确定性影响的情况,针对转速规划的伺服系统,设计了一种摩擦和转动惯量的辨识方法。采用T-S型模糊神经网络对摩擦和转动惯量进行在线辨识。将辨识得到的基于T-S型模糊神经网络的摩擦模型作为补偿控制、转动惯量用来自整定PI控制器的参数。仿真实验结果显示,设计的在线辨识方法具有良好的逼近性能,获得了满意的轨迹追踪效果。It is difficult for servo system to achieve the desired control with the influence of uncertain frictions,which also exert great unexpected impacts on motor parameter identification.Focusing on the servo system with speed planning,we design an identification method of the friction and the moment of inertia considering inherent friction.With the consideration of the strong nonlinearity of friction,T-S fuzzy neural network is employed to identify the friction and the moment of inertia online.The identified friction model based on T-S fuzzy neural network is used as compensation control while the moment of inertia is used as parameters from the tuned PI controller.Simulation results show our identification method achieves fairly good approximation performance and satisfactory trajectory tracking effect.

关 键 词：伺服系统 参数辨识 模糊神经网络 自适应控制 

分 类 号：TM341[电气工程—电机]