基于神经网络的虚拟轴机床轨迹跟踪控制方法的研究  

Research on Trajectory Track Control Based on Neural Network for Virtual Axis Machine Tool

作　　者：蓝益鹏[1] 王雷[2] 郭庆鼎[1] 

机构地区：[1]沈阳工业大学电气工程学院,辽宁沈阳110023 [2]沈阳第一机床厂,辽宁沈阳110023

出　　处：《机床与液压》2007年第12期10-11,51,共3页Machine Tool & Hydraulics

基　　金：国家自然科学基金资助项目(No.50075057);辽宁省教育厅科学研究计划资助(2004D040)

摘　　要：针对永磁直线电机驱动的六自由度虚拟轴机床提出基于神经网络的轨迹跟踪控制方法。采用神经网络来实现永磁直线电机伺服系统的位置和速度控制,将负载扰动、杆间耦合扰动等效地看成电动机模型参数的变化,通过调整权值来补偿不确定性扰动对轨迹跟踪控制的影响,保证了系统的鲁棒性。仿真实验结果表明,用该方法设计的虚拟轴机床控制系统具有良好的跟踪给定和抑制扰动的效果,从而保证机床运动协调、姿态合理。The trajectory tracking control based on neural network for 6-dof virtual axis machine tool driven by permanent magnetic linear synchronous motor was presented. The position and speed control were realized by neural network. To regard the load disturbances and coupling among links as the parameter variations of motor model, and to compensate the uncertainty disturbances by using regulation network weight, the robustness of the system was guaranteed. The results of simulation test indicate that the virtual axis machine tool with neural network controller can restraint disturbance and track of input signal, it satisfies the control requirement of motion harmony and reasonable posture.

关 键 词：虚拟轴机床 直线伺服系统 神经网络 轨迹跟踪 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]