基于神经网络的并联机床位姿误差分析与补偿研究
Research on Pose Error Analysis and Compensation of Parallel Machine Tool Based on Neural Network
孟阳阳 1陈秀梅 1彭宝营 1张美格1
作者信息
- 1. 北京信息科技大学机电工程学院,北京 100192
- 折叠
摘要
为了解决并联机床位姿误差补偿这一难题,以六自由度并联机床为研究对象,基于并联机床的运动学逆解建立了位姿误差模型,得到了并联机床驱动杆杆长误差和铰链点位置误差与并联机床动平台位姿误差的关系.铰链点位置误差可以通过调整杆长来弥补,由于驱动杆杆长误差难以测量,所以基于BP神经网络设计了一种改进的修正系统输入法对并联机床动平台的位姿误差进行补偿.仿真结果表明,经过补偿后的位姿误差明显小于补偿前的位姿误差,验证了并联机床位姿误差分析与补偿方法的可靠性.
Abstract
In order to solve the problem of pose error compensation of parallel machine tool,this paper takes 6-DOF parallel machine tool as the research object,establishes the pose error model based on the in-verse kinematics solution of parallel machine tool,and obtains the relationship between the length error of the driving rod and the position error of hinge point and the pose error of the moving platform of parallel machine tool.The position error of the hinge point can be compensated by adjusting the length of the rod.Because the length error of the drive rod is difficult to measure,an improved correction system input meth-od based on BP neural network is designed to compensate the pose error of the moving platform of the par-allel machine tool.The simulation results show that the pose error after compensation is obviously smaller than that before compensation,which verifies the reliability of the pose error analysis and compensation method of parallel machine tool.
关键词
并联机床/误差分析/BP神经网络/修正系统输入法/补偿Key words
parallel machine tool/error analysis/BP neural network/fixed system input method/compensation引用本文复制引用
基金项目
科技部重点研发课题(2021YFC2400302)
北京市教委科技一般项目(KM202011232012)
宁夏回族自治区自然科学基金(2022AAC03344)
出版年
2024
NETL
NSTL
万方数据