期刊文献+

采用支持向量机的非对称阀控液压缸模型预测控制 被引量:13

A Model Predictive Control of Asymmetric Valve Control Hydraulic Cylinders Based on Support Vector Machine and Sequential Quadratic Programming Algorithm
在线阅读 下载PDF
导出
摘要 针对非对称阀控液压缸位置控制中出现超调与由阀芯换向引起的压力冲击问题,设计了基于支持向量机与序列二次规划算法的非线性模型预测控制器。采用支持向量机拟合控制信号与活塞速度的关系得到系统的近似模型,使用序列二次规划算法求解获取控制器输出;通过增加输出约束以及满足输入约束实现每次控制过程中阀芯不换向、活塞快速到达目标位置并且无超调。数值仿真结果表明:在相对位置误差信号大于8%时,控制器输出的控制信号始终大于最大控制信号的90%,有效缩短了系统的响应时间;在保证活塞位置无超调、阀芯不换向的条件下,实现了精确位置控制,稳态误差小于0.02%。所提算法实现了多约束条件下非对称阀控液压缸系统的无超调快速控制,具有一定的理论和工程价值。 A nonlinear model predictive controller based on a support vector machine and a sequential quadratic programming algorithm is designed to solve problems of overpressure in the position control of asymmetric valve-controlled hydraulic cylinders and pressure shock caused by spool reversal.The support vector machine is used to fit the relationship between the control signal and the piston speed and to obtain an approximate model of the system,and the sequential quadratic programming algorithm is employed to obtain outputs of the controller.The valve spool is not reversed and the piston displacement quickly reaches the target position without overshoot by adding output constraints and constraints satisfying input.Numerical simulation results show that when the signal of relative position errors is greater than 8%,the control signal output by the controller is greater than 90%of the maximum control signal,which effectively shortens the response time of the system.At the same time,accurate position control is realized and the steady-state error is less than 0.02%under the condition of no overshooting of the piston position and no reversing of the valve spool.Therefore,the proposed algorithm realizes the fast control of the asymmetric valve-controlled hydraulic cylinder system without over-tuning under multi-constraint conditions,and has certain theoretical and engineering value.
作者 杨泰春 陶建峰 覃程锦 刘成良 YANG Taichun;TAO Jianfeng;QIN Chengjin;LIU Chengliang(School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)
出处 《西安交通大学学报》 EI CAS CSCD 北大核心 2020年第1期93-100,107,共9页 Journal of Xi'an Jiaotong University
基金 上海张江国家自主创新示范区专项发展资金重点资助项目(201705-XH-C1085-015) 国家重点研发计划资助项目(2018YFB1702500) 上海隧道工程有限公司专项研究科研资助项目(2017-SK-09)
关键词 非对称阀控液压缸 模型预测控制 支持向量机 序列二次规划算法 asymmetrical valve-controlled hydraulic cylinder model predictive control support vector machine sequential quadratic programming algorithm
  • 相关文献

参考文献2

二级参考文献27

共引文献568

同被引文献120

引证文献13

二级引证文献36

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部