直线伺服系统模型参数辨识及其实验研究  

Parameter Identification and Experimental Research of a Linear Servo System

作　　者：陶大锦[1] 柯大观[1] 袁柳斌[2] 朱丹峰[1] 

机构地区：[1]温州医学院,温州325035 [2]西安交通大学,西安710049

出　　处：《中国机械工程》2013年第7期915-921,共7页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(11005081);浙江省教育厅资助项目(Y201016712);温州医学院科研发展基金资助项目(QTJ09003)

摘　　要：为了使伺服驱动系统具有较好的控制性能,对系统建模与辨识、系统性能进行了研究。传统方法建立的直线伺服系统数学模型误差大,为了求出精确的系统模型参数,将基于最小二乘的阶跃辨识法应用在直线伺服系统的模型参数辨识中,并计算出实验数据与仿真数据的拟合误差(0.0018)和相关系数(0.9970)。仿真与实测结果表明,该模型辨识算法可行,对输出测量噪声不敏感,鲁棒性好。采用极点配置法设计了位置控制器,并对其进行了实验研究。通过合理配置系统极点的位置,显著地减小了摩擦力对系统性能的影响。In order to improve the control performance of a linear servo system, system modeling and identification, control performance of the system were studied. The error of the modeling results worked out by traditional method was large. In order to work out the model parameters, a step re- sponse identification with least square method was applied to build the model. The fitting error and correlation coefficient between experimental and simulation data were 0. 0018 and 0. 9970 respectively. The simulation results show that the model identification is feasible and the resulting parameter esti- mation is insensitve to measurement noise. The position controller was designed by pole placement method. Then an experimental investigation was conducted to study the control performance of the position controller. The Influence of friction on system performance can be reduced dramatically by pole placement method.

关 键 词：直线电机 参数辨识 控制性能 数学模型 最小二乘 

分 类 号：TM383.4[电气工程—电机] TP275[自动化与计算机技术—检测技术与自动化装置]