Research And Design Of Self-tuning Feedforward Controller For Motion Control System

Motion control systems are widely used in all areas of modern manufacturing.The linear servo system is a motion control system which can directly output linear feed motion.Because of its simple structure,high positioning accuracy,easy adjustment and control,it is widely used in applications where high-precision linear motion is required.Because high-precision motion control systems often have external disturbances,uncertain changes in model parameters,etc.,it is necessary to study and design the motion controllers and control algorithms for the motion control system.The motion control system is often required to achieve full tracking of a given signal.Firstly,the motion control system is divided into a minimum phase system and a non-minimum phase system depending on whether the motion control system has a non-minimum phase zero.For the minimum phase system,this paper introduces the feedback controller and feedforward controller of two-degree-of-freedom control strategy.After the feedforward controller structure is obtained,a model-based and data-based Newton iterative learning control algorithm is proposed to update the parameters of the feedforward controller to improve the trajectory tracking performance of the system.Moreover,the order of the feedforward controller can be adjusted according to the order of input trajectory and the accuracy requirements of the system,so as to meet the requirements of different motion control occasions.Then,the non-minimum phase system is unstable because there are the non-minimum phase zeros.And we can't use the precise inverse of the system to design the feedforward controller.Therefore,non-minimum phase zero ignore technology,zero phase error tracking control technology and the zero-amplitude error tracking control technology are used to design the feedforward controller in this paper.The three model inverse control techniques developed can improve the instability of the feedforward controller caused by the non-minimum phase zero.And,the actual motion control system often has a series of problems such as external disturbance,which has a great influence on the trajectory tracking accuracy and the stability of the system.In this paper,a feedforward controller parameter self-correcting control structure is introduced to solve the problem of non-minimum phase zero variation in a non-minimum phase system and to realize automatic correction of feedforward controller parameters.Finally,we introduce the hardware platform and software platform of the permanent magnet synchronous linear motor servo system used in this experiment.For the minimum phase system,modeling and simulation are carried out on the MATLAB platform,and the parameter optimization experiment of the minimum phase system feedforward controller is carried out to verify the rationality of the feedforward controller structure and the feasibility of the parameter optimization algorithm.And the simulation results illustrate that the application of three model inversion techniques in the non-minimum phase system and the correction effect of the self-correcting control structure algorithm when the parameters of the non-minimum phase system change are verified.