Font Size: a A A

Research On High-speed And High-precision Motion Control Based On Optimal Iterative Learning

Posted on:2022-10-05Degree:MasterType:Thesis
Country:ChinaCandidate:R YuanFull Text:PDF
GTID:2518306548962039Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Iterative learning control is generally used in systems that perform repetitive tasks.The error signal is used to correct the control signal,thereby gradually reducing the output error of the system,and finally achieving complete tracking of the desired trajectory.With the development of science and technology,many scholars have proposed different types of iterative learning control algorithms,mainly including P-type,PD-type,H_?-type,and optimal iterative learning control.Among them,the optimal iterative learning control algorithm is composed of optimal control theory and iterative learning control.It has the advantages of fast transient response,monotonic convergence,simple controller design,and convenient algorithm implementation.It has been successfully applied to industrial robots,although there have been a lot of research results in high-speed and high-precision motion control situations such as CNC machine tools and lithography machines,there are still some places that need to be supplemented and improved.This article first introduces the feedback and feedforward two-degree-of-freedom control structure.PID feedback control is used to ensure the stability of the system.On this basis,the iterative learning controller is designed according to the optimal control theory.In addition,parameters such as weighting matrix and Lagrange operator are introduced into the algorithm to enhance the stability and robustness of the algorithm,and at the same time,the algorithm can flexibly adjust the convergence speed.Through theory,simulation and experiment,the effect of weighting matrix and Lagrange operator on the algorithm is comprehensively analyzed.Secondly,it analyzes the characteristics of point-to-point motion,and introduces the execution time period and adjustment time period window matrix on the basis of the optimal iterative learning control to suppress the residual vibration.The introduction of the window matrix can not only effectively suppress the residual vibration of the high-speed motion system and improve the positioning accuracy of the system,but also can change the control target of the algorithm by adjusting its size,and improve the flexibility of the optimal iterative learning controller design.After that,the influence of the time-varying characteristics of the system on the traditional optimal iterative learning control is analyzed,and two adaptive optimal iterative learning control algorithms are proposed.First,a data-driven adaptive optimal iterative learning control is proposed.The optimal iterative learning controller is updated by identifying each row of non-zero elements in the nominal model of the system,and its convergence and tracking performance are analyzed.Second,an adaptive optimal iterative learning control algorithm based on least squares identification is proposed.The unit impulse response of the system is identified in the iterative domain,the controller is updated according to the identification value,and the performance index criterion is introduced.And Butterworth filter to improve the identification accuracy of the algorithm.Both algorithms can effectively cope with the time-varying characteristics of the system,without the need to build a system parameter model structure,and make up for the shortcomings of the optimal iterative learning control algorithm.Moreover,the second algorithm has a faster convergence speed and a smaller amount of calculation.Finally,design a reasonable experimental program and build an experimental platform based on the brushless DC servo motor.The above algorithm is applied to the actual servo system control,and the correctness of the algorithm is verified through a large number of simulations and experiments.
Keywords/Search Tags:Optimal iterative learning, High-speed and high-precision motion control, System identification, Data-driven, Least squares
PDF Full Text Request
Related items