Font Size: a A A

Pid Control System Based On Bp Network For Winding Machine Parameters Study

Posted on:2006-05-28Degree:MasterType:Thesis
Country:ChinaCandidate:H J TuFull Text:PDF
GTID:2191360152982147Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
Resin composite components are widely applied in advanced technology in space field. Winding, the main shaping technique, is performed by winding machines. In the process, the control of the stability of the tension is one of the crucial techniques ensuring the component quality of the composite materials. The first purpose of applying tension is to. Suitable and stable tension can increase the winding component ability undertaking the internal pressure and the ability against fatigue. Thus the winding machine of good quality must be equipped with nice tension controller.One type of tension control system are researched and manufactured by this project. This system with industrial control computer as the core, magnetic particle clutch as the actuator. The detailed analysis is made on its principles, static and dynamic properties of the actuator and the selected parameters of it's main components. On the basis of study on unwinding scroll of the tension system, the mathematical model of tension control system is established. Meanwhile, the PWM circuits suitable for driving are developed to driver the actuator.In industrial control, The PID control is widely used because it need not relying on the accurate mathematics model, but when the change which input way or environment makes the form interfered change, three pieces of control coefficient (P , I , D ) of PID control system need exactly making artificially again, it is difficult to make the satisfactory control result. While the neural network control system has the characteristics of robustness, fault tolerance and parallel operation structure, independence on the mathematical model of the controlled object, and the adaptive to any indefinite system, the theory of the neural network is applied on the PID system. It has testified by experiment that the principle of the precise tension control system designed on the basis of neural network theory is right, the run is reliable and the deviation of tension is small, thus reach the required property parameter of the system.This text has also simply introduced the realization of the control system of the pressure and control system of temperature.The function of the software system designed is described respectively.
Keywords/Search Tags:winding machine, tension control, neural network, magnetic particle clutch, serial communication
PDF Full Text Request
Related items