交流伺服控制系统的刚性自适应补偿技术  

Rigidity Adaptive Compensation Technology of Alterating Current Servo Control System

作　　者：周延超 马殿光[1] 张宁云 唐厚君[1] ZHOU Yan-chao;MA Dian-guang;ZHANG Ning-yun;TANG Hou-jun(Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学,电子信息与电气工程学院,上海200240 [2]上海交菱数控科技有限公司,上海200000

出　　处：《电力电子技术》2021年第4期34-38,共5页Power Electronics

摘　　要：为解决交流伺服控制系统在数控加工中由机床振动导致的定位精度不高的问题,提出一种基于特征建模、系统辨识和线性二次型调节器(LQR)的交流伺服控制系统刚性自适应补偿方法。该方法在线辨识得到机床的特征模型,利用状态观测器和LQR设计全状态反馈控制律,从而可有效抑制机床弱刚性对位置闭环控制的干扰。在系统辨识方面,提出一种基于特征值分析的扩展变步长最小均方误差算法(EVLMS),并与归一化最小均方误差算法(NLMS)和带遗忘因子递推最小二乘算法(RLS)进行了比较。分析和实验表明,与传统的比例积分微分(PID)位置闭环控制相比,此处提出的自适应补偿技术能有效抑制假性振动对位置闭环控制的干扰,可以提高交流伺服控制系统的动态性能和稳态精度。In order to solve the problem of low positioning accuracy caused by machine tool vibration in computer numerical control machining of alternating current servo control system,an adaptive compensation method of alternating current servo control system rigidity based on characteristic modeling,system identification and linear quadratic regulator(LQR)is proposed.This method obtains the characteristic model of the machine tool online,and uses the state observer and LQR to design the full state feedback control law,which can effectively suppress the interference of the weak rigidity of the machine tool on the position closed-loop control.In terms of system identification,a extended variable step size least mean square error algorithm(EVLMS)based on eigenvalue analysis is proposed,and compared with normalized least mean square error algorithm(NLMS)and recursive least squares algorithm(RLS)with forgetting factor.Analysis and experiments show that,compared with the traditional proportional integral differental(PID)position closed-loop control,the proposed adaptive compensation technology can effectively suppress the interference of fake vibration on the position closed-loop control,and can improve the dynamic performance and steady-state accuracy of the alternating current servo control system.

关 键 词：伺服控制系统 刚性 自适应补偿 特征建模 

分 类 号：TN820.3[电子电信—信息与通信工程]