期刊
MECHANISM AND MACHINE THEORY
卷 175, 期 -, 页码 -出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.mechmachtheory.2022.104954
关键词
Dual-flexible servo system; Sliding mode control; Neural network; Nonlinear terms
资金
- National Natural Science Foundation of China [51875092]
- National Key Research and Development Project of China [2020YFB2007802]
- Natural Science Foundation of Ningxia Province [2020AAC03279]
- Fundamental Research Funds for the Central Universities [N2103025]
This paper proposes a sliding mode control strategy based on neural network compensation to control a variable-length dual-flexible servo system. By considering the 2D deformation of the flexible load, the mechanical behavior of the servo system is described more accurately. The experimental results demonstrate that the control strategy can significantly reduce the tracking error.
A variable-length dual-flexible servo system is a complex nonlinear system that consists of a servo motor, a flexible joint, and a variable-length flexible load. In order to describe the mechanical behavior of the servo system more accurately, the two-dimensional (2D) deformation of flexible load needs to be considered. The coupled nonlinear terms, friction torque, and time variation of parameters in the dynamic equations constitute the uncertain part of the servo system. These mentioned factors will cause fluctuations in the flexible load's rotation angle. In this paper, a sliding mode control strategy based on neural network compensation is proposed to control the servo system. Firstly, the dynamic equations of the dual-flexible servo system considering the 2D deformation are established according to Lagrange's theorem. Then the control law of the sliding mode controller based on neural network is designed. The adaptive law of weight coefficients in the neural network is designed by the Lyapunov stability theorem. Finally, the numerical simulation analysis and control experiments are carried out. The experiment results prove that the control strategy can reduce the tracking error by 43.6%.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据