Resonance Detection And Suppression Of Permanent Magnet Synchronous Servo System

The mechanical transmission device in the permanent magnet synchronous servo system is not an ideal rigid body and has a certain elasticity.In order to obtain high response,the servo control often increases the proportional gain of the speed regulator,which will cause mechanical resonance and make the entire servo system out of adjustment.Therefore,it is of great significance to adopt corresponding strategies to suppress resonance.In this paper,the ball screw is used as the mechanical transmission device of the permanent magnet synchronous servo system,the resonance problem caused by the coupling and the screw is analyzed,and the notch filter is used to filter out the resonance frequency in the speed loop,and finally the resonance suppression is achieved.The paper first briefly introduces the mathematical model of the permanent magnet synchronous motor in the synchronous rotating coordinate system and the basic structure of vector control.Secondly,the multi-mass model of the ball screw elastic system was established,and the model was simplified to be a dual-mass model.Then,based on the dual-mass model,the mechanical resonance generation was analyzed using bode diagram and description function method The reason for this is that when the resonance occurs,the vibration frequency of the output signal of the speed regulator is approximately equal to the resonance frequency of the system,thus laying a theoretical foundation for the detection of resonance frequency in the subsequent chapters.Subsequently,in view of the shortcomings of the traditional FFT resonance frequency detection method,such as poor real-time performance,this paper proposes an ANF-based resonance frequency detection method.The resonance frequency of the system is obtained by real-time vibration frequency estimation of the output signal of the speed regulator.This improves the detection speed of the resonance frequency of the system.The principle and stability analysis of the ANF algorithm are introduced,and the derivation process of the digital realization of the algorithm is given.At the same time,for the problem of nonconvergence of frequency estimates caused by the traditional bilinear transform discretization method,an improved bilinear transform method is proposed to effectively solve this problem.Finally,the ANF algorithm is applied to the resonance frequency detection of the digital servo system,and the simulation verifies its effectiveness.In addition,in order to suppress resonance,this paper uses a resonance suppression strategy based on a notch filter,briefly introduces the structural characteristics of the traditional notch filter,and for the shortcomings of the traditional notch filter,an improved notch filter is proposed And connect the notch filter between the output of the speed regulator and the given current loop to filter out the resonant frequency in the speed loop and finally suppress the resonance.Finally,an experimental platform using ball screw as the mechanical transmission device of permanent magnet synchronous servo system was built,and the feasibility and effectiveness of the resonance frequency detection method and resonance suppression strategy in this paper were verified through experiments.