High Precision Control Motheds Of Hexaglide Virtual Axis Machine Tools

Although the structure of virtual axis machine tool is very simple, it can process the complex curved surface which can not be processed by general data control machine tool. But for the characteristics of the structure of virtual axis machine tool, there are strong mechanic coupling among every axis. Therefore, the motion of every axis will effect other axises. and the control method become relatively more complex than normal machine tools. How to realize the high precision control is an important problem must be solved. Our task is to study the high precision control of hexaglide virtual axis machine tool.Line motor is adopted as driving machine in hexaglide virtual axis machine tool, so the control object of the servo system in the machine tool is line motor. Here line permanent magnetic AC motor is used as the driven equipment of the hexaglide virtual axis machine tools.To satisfy high performance, many causes which will destroy the performance of the servo system are considered carefully in the system where line permanent magnetic servo motor is control object: the mechanic coupling disturbance among every axis, the mass change of the mover of line permanent magnetic motor, the uncertainties of the model parameters made by the change of the friction coefficient, the unstructured uncertainties caused by ignoring the electromagnetic time constant and the thrust fluctuations aroused by the varieties of the magnetic density distributions and the time harmonic. In order to reduce the effects on system brought by the outside disturbances and model errors, to improve the servo performances and to guarantee automatically the robust stability , the H_∞ robust control law is app led to the servo system in this paper and combined with the disturbance observer, therefore the control precision can be assured by using the data which is observed by the observer-to complement the disturbance of the environment at any moment.At the same time a new type of neural network thrust compensator is proposed here. Forthe clear physical meaning of its weight, and the little number of the en-line adjusted weight this kind of neural network have the fast response performance which other neural networksdon't have. Therefore this neural network can give the system real-time compensation, and the proposed control scheme has a rigorous theoretic base and has been validated by the theory and experiment.When uncertainties of the parameter, unmodeled dynamics and bounded perturbation 'exists simultaneously, the precision control of the control object which have many uncertainties can be realized in the servo system by using H∞ law analysis, IP position control design, and neural network.. The result of simulation indicates that this control scheme have strong robust and good control effect.