Adaptive Repetitive Learning Control Of Permanent Magnet Synchronous Motor Systems

As one of the most commonly used actuators of automation equipment,permanent magnet synchronous motors(PMSM)play an indispensable role in modern industry.In practical applications,permanent magnet synchronous motors often need to perform periodic tracking tasks.In this case,there will be many periodic disturbances in the system.By using the operating data of the previous cycle to modify the control input of the current cycle,repetitive learning control can improve the tracking accuracy of the system's periodic trajectory or eliminate the effects of periodic disturbances.In this dissertation,the adaptive repetitive learning control of the permanent magnet synchronous motor systems is studied to achieve high-precision tracking of the periodic dedired trajectory by the system The main research contents include:.(1)The mathematical model of parametric uncertain permanent magnet synchronous motor system is established,and then design the adaptive robust controller.Furthermore,in view of the periodic characteristics of parameter uncertainty in the system,an adaptive repetitive learning control schema is proposed.By designing the repeatitive learning law to estimate and compensate the periodic uncertainty,the influence of periodic disturbance is eliminated,and the tracking performance of the system on the periodic reference trajectory is enhanced.In addition,the bounddness of the learning law could be guaranteed by introducing a saturated learning mechanism.(2)The mathematical model of the nonparametric uncertain permanent magnet synchronous motor system is proposed.According to the periodic characteristics of the reference trajectory,the nonparametric uncertainty is divided into two parts: periodic uncertainty and nonperiodic uncertainty.The periodic uncertainty is involved in an unknown desired control input,and a fully saturated repetitive learning law is developed to ensure that the estimate of the unknown desired control input is confined with a prespecified region.Then,the nonperiodic uncertainty is converted into the parameterized form,and an adaptive update law is designed to compensate for it.The designed adaptive repetitive learning controller can achieve high-precision tracking of the periodic reference trajectory of the system,and since it does not include a sign function,the chattering problem of control signal is effectively avoided.On this basis,the system model will be extended,and the adaptive repetitive learning control design method is extended to a class of nonparametric uncertain systems.(3)Based on the digital signal processor DSP28335,an experimental platform for permanent magnet synchronous motors is established to verify the control schema proposed in the paper.The experimental results show that,through repetitive learning estimation and compensation of the periodic desired control input,the adaptive repetitive learning control schema proposed in this dissertation can achieve high-precision tracking of the periodic reference trajectory...