Learning Feedforward Control Algorithm And Resonance Suppression Of High Speed Linear Servo System

Linear servo system is a direct drive to the servo system,there is no middle of any drive link.It has many features,like direct output linear motion,with a simple structure,suitable for high-speed and high-acceleration movement,easy to adjust and control features,which are widely applied in high-speed high-precision CNC systems,semiconductor manufacturing and precision instruments and other fields.In order to exert its high-speed high-precision characteristics,constantly research the control algorithm and design of the control controller is needed.This paper first introduces the two-degree-of-freedom control strategy of the linear servo system,the feedback control of it can only keep the robustness of the system for the feedback control,but can not improve the trajectory tracking performance of the servo system.A linear feedforward controller is proposed,and the feedforward controller is parameterized.The feedforward controller can adjust its order according to the modeling precision of the servo system and the order of the input trajectory planning to achieve different servo control accuracy.In order to solve the influence of the linear time delay characteristic of the linear servo on the trajectory tracking performance,the influence of the linear servo delay characteristic on its trajectory precision is analyzed on the basis of the feedforward and feedback composite controller.Time delay controller is designed.The delay controller can be designed as an advance link control which can not be realized when the time-delay controller is located in the feed-forward channel.The time-delay controller is divided into two parts,the ideal trajectory instruction follows the integer multiple of the servo cycle delay and feedforward control signal filtering delay to achieve the purpose of compensating for the time lag.Secondly,with the increasing linear servo system bandwidth,the influence of the flexible characteristics of the linear servo system becomes more and more obvious,which leads to the mechanical resonance.In this paper,the "spring-damping" model is introduced to analyze the resonance characteristics of the linear servo system.Passive resonance suppression scheme is selected,the notch filter is added to the controller,and the effect of the filter on mechanical resonance suppression is verified by matlab simulation.The iterative learning control method of the feedforward controller,the delay controller andthe notch filter is used to optimize the parameters of the controller and improve the performance of the servo system in view of the repeatable motion characteristics of the linear servo system.The gradient descent method,the Newton method and the secant method are introduced respectively.The iterative method is deduced and the controller parameters are added to derive the iterative formula which can be used in the simulation experiment.Finally,the hardware and software platform of the linear servo system of the laboratory are introduced.On the basis of iterative method,the iterative optimization simulation and experiment of the three controllers are carried out to verify the effect of the controller and the iterative optimization method.