A Sparse Sampling Modeling Method And Self-tuning Control For Brushless DC Motor Driven System

In this paper,a parameter identification method combining Kalman filter with least squares method is proposed to solve the difficulty of precise modeling of sparsely sampled brushless DC motor drive system.For the new grey box model,a new switching self-tuning control scheme is proposed,which solves the difficulty of poor control effect of general self-tuning control system in initial operation.Establishing accurate mathematical model is the first step to realize high performance control system.Empirical model is usually used when the parameters of controlled object model ai*e unknown.However,the lack of enough precise data brings difficulties and challenges to the modeling process.In the absence of speed sensor,this paper uses Kalman filter to estimate accurate speed data,and uses recursive least squares method to identify parameters online,and obtains the mathematical model of brushless DC motor drive system.Then a new switching self-tuning controller is designed according to the mathematical model.Experiments show that the proposed modeling method can establish an accurate mathematical model for the sparsely sampled brushless DC motor drive system.The experiment also proves that the new switched self-tuning controller has better control effect than the unmodified self-tuning controller.The contents and innovations of this paper are as follows:1.This paper independently designed an automatic equipment platform for automobile wire harness winding process.The equipment platform can cariy out motor drive experiment and data transmission.In the development of equipment platform,the innovations of engineering technology are:a method of using timer to implement simple operating system is proposed,and multithreading programming can also be realized for low performance CPU.An improved environmental protection algorithm with temperature lag is proposed?which can effectively restrain the frequent action of the actuator.A new communication fomiat is proposed,which avoids the packet error caused by fixed frame head and reduces the traffic.2.Refining academic problems from practical engineering problems.In engineering practice,because of installation space or economic performance,no speed sensor is used in BLDCM drive system,and the on-line running data cannot be collected for system identification of dynamic control system,so it is impossible to establish accurate transfer function or state space model.The engineering problem is refined into a sparse sampling modeling problem for brushless DC motor drive system.3.Aiming at the problem of sparse sampling modeling of brushless DC motor drive system,a parameter identification method combining Kalman filter and least square method is proposed in this paper.Combining the speed data calculated by Hall position sensor with the speed data estimated by the auxiliary model,Kalman filter is used to estimate the optimal speed data,and then the least square method is used to identify the parameters and get the model of the controlled object.Experiments show that the proposed modeling method can improve the modeling accuracy,and the mean square error of the model will be reduced by more than three times.4.For the new grey box model,aiming at the problem of poor control effect of self-tuning controller in initial operation,a new switching self-tuning control scheme is proposed,which makes use of the advantages of PID controller,pole placement controller and self-tuning PID controller respectively,and has excellent control effect.Expek riments show that the proposed switching self-tuning controller can improve the performance of the system and reduce the adjustment time of dynamic response to 1/2 of the unmodified self-tuning controller.