基于滑窗滤波器的改进MRAS转动惯量辨识  

Improved Moment of Inertia Identification of MRAS Based on Load Disturbance Observation

作　　者：陈冠廷 魏海峰[1] 闵琦 CHEN Guanting;WEI Haifeng;MIN Qi(School of Automation,Jiangsu University of Science and Technology,Zhenjiang 212000,China)

机构地区：[1]江苏科技大学自动化学院,镇江212000

出　　处：《组合机床与自动化加工技术》2024年第4期110-114,120,共6页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(51977101);江苏省省重点研发计划产业前瞻性与共性关键技术重点项目(BE2018007);镇江市产业前瞻性与共性关键技术重点研发项目(GY2019033)。

摘　　要：交流伺服电机在负载突变时,传统模型参考自适应(model reference adaptive system,MRAS)不能同时满足辨识精度和收敛速度两方面的需求,鉴于此,提出一种基于滑窗滤波器的改进MRAS转动惯量辨识策略。首先,利用曲率模型(curvature model,CM)对电机的负载扰动进行实时估测;其次,对传统MRAS法进行改进,基于滑窗滤波器原理设置变增益机制的判断依据,使负载突变时系统自动调整自适应率中的增益系数,确保转动惯量被快速、精准的辨识;最后,通过仿真和搭建实平台对所提方法进行验证,结果证明此方法不仅具备较高的辨识精度,同时收敛速度快,很好的克服了传统方法的不足,表现出优异的动态性能。The traditional model reference adaptive system(MRAS)cannot meet the requirements of identi-fication accuracy and convergence speed at the same time when the load of AC servo motor changes rapidly.In view of this,an improved MRAS moment of inertia identification strategy based on sliding window filter is proposed.First,the curvature model(CM)is used to predict the load disturbance of the motor in real time,and then the traditional MRAS method is improved.Based on the sliding window filter principle,the variable gain mechanism is set up to determine the basis,so that the system automatically adjusts the gain coefficient in the adaptive rate when the load changes,ensuring that the moment of inertia is quickly and accurately iden-tified.Finally,the proposed method is verified by simulation and real platform construction.The results show that the proposed method not only has high identification accuracy,but also has fast convergence speed,which overcomes the shortcomings of traditional methods and shows excellent dynamic performance.

关 键 词：交流伺服电机 模型参考自适应 负载扰动 惯量辨识 滑窗滤波器 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]