Adaptive Control Reserch Of Pneumatic Force Control System

Pneumatic force servo system is widely used in practical engineering, usuallyPID control strategy.PID controller’s robustness is poor, if load or otherparameters changes, can’t guarantee the system performance. Due to the complexityof the Pneumatic force servo system, the adjustment of PID parameters is toocomplicated to master for the general users, the application of pneumatic forceservo system is limited. To solve this problem this paper studies the adaptivecontrol. Adaptive control is one important kind of pneumatic force servo systemcontrol strategy, mainly used to solve the problem of poor robustness. Adaptivecontrol technology is successfully used in many fields, such as the optical trackingtelescope, chemical industry, metallurgy, machinery processing, etc. This paperstudies adaptive control technology of pneumatic force servo system.At first, this paper established a nonlinear simulation model of pneumaticforce servo system and its linearization, used the actual system parameters, get thetransfer function of the system.To study the system characteristics, the nonparametric identification andparameter identification is studied. Firstly identify signals-inverse M sequencesignal is designed, then studied the identification data acquisition and processing,using H1algorithm to complete the nonparametric identification, designed therecursive identification program to identify system parameter. At end, identified thenonparametric model and parameter model of the simulation model. Thenonparametric model and parameter model with similar frequency response, the fitof the parameter model output and the simulation model output is82%.According to the characteristics of the pneumatic force servo system, designedthe nonlinear PID controller,generalized minimum variance self-tuning controllerand pole placement self-tuning controller. The simulation results show that theconvergence time of online identification is about0.05s and the system usedadaptive controller can keep good performance in variable working conditions,have stronger robustness than PID control system.In the end, this paper constructs the control system model and designs the program of the control system by using the software Matlab/Simulink and XPCtarget to realize the real-time control of the test system. Have the test systemdebugged and identified.According to the control strategies proposed above, aseries of experiments is carried out. The experimental results show that theconvergence time of online identification is about0.05s, system used adaptivecontroller have stronger robustness than PID control system.