Accurate inertia identification method of machine tool feed drives by considering the influence of current loop dynamics and friction. (September 2022)

• Record Type: Journal Article
• Authors: Zhan, Chengpeng; Yang, Xiao; Lyu, Dun; Zhao, Wanhua; Chen, Yaolong
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 236:Number 8(2022)
• Page Start: 1447
• Page End: 1463
• Publication Date: 2022-09
• Subjects: Current loop dynamics -- inertia identification -- least squares method -- instrumental variable method -- machine tool feed drives -- Stribeck's friction; Mechanical engineering -- Periodicals; Automatic control -- Periodicals; Systems engineering -- Periodicals; 621.3
• DOI: 10.1177/09596518221100290
• Languages: English
• ISSNs: 0959-6518

Abstract

Accurate identification of the moment of inertia is the basis of modeling, simulation, and controller design of machine tool feed drives. However, insufficient consideration of colored noise factors (such as current loop dynamics, Coulomb's friction, Stribeck's friction, and nonlinear damping) will lead to inaccurate inertial identification results. This article proposes a new accurate offline identification method to accurately identify the equivalent inertia of machine tool feed drives. It is based on the least squares method and the instrumental variable method, and the equivalent time constant of the current loop, Coulomb's friction parameters, Stribeck's friction parameters, and the nonlinear damping parameter can be identified simultaneously with the inertia. First, a discrete transfer function of feed drives that considers the first-order dynamics of the current loop, Coulomb's friction, Stribeck's friction, and nonlinear damping is established. Then, inertia and the equivalent time constant of the current loop, Coulomb's friction parameters, Stribeck's friction parameters, and the nonlinear damping parameter are identified simultaneously based on the least squares method. Third, the instrumental variable method is used to correct the parameters identified by the least squares method. Finally, the inertia identification experiments are carried out on a ball screw feed drive system under different types and amplitudes of input signals. The experimental results show that the proposed inertia identification method can effectively improve the accuracy of inertia identification and its robustness to the input signal amplitude.