深水清淤机械臂阀控液压油缸的加速干扰自适应滑模控制方法
Accelerated-interference adaptive sliding mode control method for the valve-controlled hydraulic cylinders of a deep-water dredging manipulator
樊帆 1郑皓 1史浩东 1彭赛锋1
作者信息
- 1. 长沙矿冶研究院有限责任公司 深海矿产资源开发利用技术国家重点实验室,湖南 长沙 410012
- 折叠
摘要
针对深水清淤机械臂阀控非对称油缸因高阶非线性特性以及水下大干扰工作环境导致的控制困难问题,本文提出一种加速干扰自适应滑模控制方法.通过整理系统的高阶非线性模型,建立液压油缸的分层控制模型.使用反步法设计了滑动模态指数收敛的滑模控制器,在此基础上提出了一种加速的干扰自适应方法,并使用李雅普诺夫稳定性理论验证了该复合控制方法的稳定性.通过Simscape平台进行多物理域仿真验证,仿真结果表明:所设计的加速干扰自适应滑模控制方法能实现快速准确的位置跟踪,相比于比例积分微分控制具有响应速度快、超调量小的特点,相比于普通干扰自适应滑模控制方法具有稳态误差更低,干扰估计速度更快的特点.
Abstract
With the goal of overcoming the difficulties faced by a valve-controlled asymmetric oil cylinder of a deep-water dredging manipulator due to the characteristics of high-order nonlinearity and underwater working environment with large disturbance,this paper proposes a new adaptive sliding mode control method with accelerated disturbance.We established a hierarchical control model of a hydraulic cylinder by sorting out the high-order nonlinear model of the system.Using the backstepping method,a sliding mode controller was designed with a convergent sliding mode exponent,after which an adaptive method with accelerated disturbance was proposed based on this method.Furthermore,we verified the stability of this compound control method based on the Lyapunov stability theory.Using the Simscape platform,simulations were conducted in multiple physical domains.The results showed that the proposed accelerated disturbance adaptive sliding mode control method can achieve fast and accurate position tracking.Moreover,compared with the proportional-integral-derivative(PID)control,this mode has such characteristics as fast response and small overshoot.Finally,compared with the adaptive sliding mode control method with common disturbance,it possesses the characteristics of lower steady-state error and faster interference estimation speed.
关键词
自适应控制/滑模控制/机械臂/清淤机械/阀控液压油缸/非线性系统/神经网络/多物理域仿真Key words
adaptive control/sliding mode control/mechanical arm/dredging machinery/valve controlled hydraulic cylinder/nonlinear system/neural network/multi-physical domain simulation引用本文复制引用
基金项目
湖南省科技重大专项计划项目(2018SK1010)
国家重点研发计划(2016YFC0401607)
出版年
2023
国家科技期刊平台
NSTL
维普
万方数据