【作者】 赵洁；

【导师】 王从庆；

【作者基本信息】 南京航空航天大学 ， 模式识别与智能系统， 2004， 硕士

【摘要】 运动控制是自动控制的重要分支，由于它能够实现对运动轨迹与运行速度的精确控制要求，因此运动控制系统在各类控制工程中有着广泛应用前景。远程控制是本地计算机通过网络系统实现对远端的生产或者实验过程的监视和控制，实验者从网络上可以看到直观、可信的实验信息，对校园网络实验教学的发展有着重要的意义。 本课题针对双轴运动控制系统的远程网络控制，将网络技术与运动控制技术结合起来开发远程网络运动控制系统。 本课题以提高双轴运动控制系统控制精度为目的，从单、双轴两方面进行分析。对于单轴系统首先利用美国DeltaTau公司的可编程多轴控制器PMAC自带的控制软件PEWIN进行PID参数的整定以及前馈补偿环节的参数整定，然后分析系统存在的非线性摩擦力，利用遗传算法寻优摩擦力补偿环节的参数，达到补偿摩擦力的效果，有效地提高了系统的控制性能。对于双轴系统设计了一个轨迹为圆的插补器，采用基于遗传算法—神经网络的在线PID参数调整算法，减小了系统的跟踪误差，并分析双轴之间的轮廓误差，设计出基于RBF神经网络的综合控制器，经过仿真实验证明，本文采用的综合控制器能有效地减小系统的跟踪误差和轮廓误差，提高系统的控制精度。 最后运用网络数据库原理及SQL Server数据库编程技术，实现了网络数据库，并利用双层防火墙技术和数据库存取控制技术来实现系统的网络安全，使得双轴运动控制系统可供校园网内师生及研究人员实验和观察。更多还原

【Abstract】 Motion control is an important branch of automatic control, and because it can realize the request of precise control for motion track and moving velocity, it has got extensive application foreground. Remote control system is to supervise the remote production or experiment process, implemented by local computer through Internet. The experimentation personnel can get the results through Internet directly, so remote control has important meaning in the development of experiment teaching through campus networks.This task is to control the two-axis motion system through networks, and net technology and motion controlling technology are used to explore remote control system.In order to improve motion accuracy of the two-axis motion system, the analysis is made in two aspects-single axis and two axis. As for single axis, PEWIN, the software given by the Programmable Multi-Axis Controller (PMAC), is used to find the best parameters of the PID controller and feedforward compensation controller. Then, the nonlinear friction in the system is compensated by the friction compensation controller with Genetic Algorithm(GA). The compensation controller can improve the system performance effectively. In two axis system, a circle interpolator is designed, and a new method based on GA-Neural Networks is proposed to adjust PID parameters online to reduce the tracking error. Then, an integrated controller based on RBF Neural Network(RBFNN) is designed to reduce the contouring error of the two axis. Experimental results show that the proposed integrated controller significantly improved accuracy in both tracking and contouring.At last, network database is established by using network database theory and SQL Server programming technology. Two-layer firewall technology and database access control technology are used to ensure network security, so the students, teachers and researchers all over school can supervise and control the two-axis motion system in the Internet.更多还原