Research On Strategy Of Speed Vibration Suppression In AC Servo System

The AC servo system plays an irreplaceable role in the traditional industrial and high-tech fields,and its performance directly determines the actual performance of CNC machine tools,robots and other industrial equipment.Industrial equipment often use transmission shafts,couplings and other devices to connect the motor and load.These connecting devices are not ideal rigid bodies,and they are prone to mechanical resonance,which limits the improvement of the dynamic performance of the servo system.In order to solve these problems,this paper will study the resonance characteristics identification and vibration suppression strategy of AC servo system.Starting from the vector control of Permanent Magnet Synchronous Motor and the resonance characteristics of the dual inertia system,the paper models the dual inertia servo system by mechanism method,analyzes the open-loop and close-loop resonance characteristics of the system by frequency domain analysis method,classifies and sums up the low,medium and high frequency resonance problems,which lays the foundation for the research of the vibration suppression strategy.In order to realize the frequency characteristics identification of the servo system,an offline system identification method is designed with pseudo-random sequence as excitation,this method uses singular value decomposition and model transformation to obtain the low order open-loop transfer function of the dual inertia servo system.In addition,this paper also applies an improved off-line identification method of inertia,which optimizes the identification for the dual inertia system and improves the identification accuracy.In order to realize the on-line identification of vibration frequency,this paper proposes an on-line frequency identification method based on the second-order generalized integrator.Through the improvement of adaptive rate and discrete method,the fast and accurate identification of vibration frequency is realized.In order to effectively solve the low,medium and high frequencies resonance problems of the dual inertia servo system,this paper proposes corresponding suppression strategies for different vibration situations.For the low-frequency resonance problem,the extended state observer is designed to suppress the resonance.Compared with the traditional disturbance observer,its disturbance rejection ability and dynamic performance are significantly improved.For the medium-frequency resonance problem,this paper proposes a vibration detection method and vibration suppression method based on the speed observer.Firstly,a speed observer is used to evaluate the vibration of the system in real time and determine the entry time of the vibration suppression method,and then extract the vibration information of the fixed frequency band and transform the speed feedback loop to achieve the suppression of the mediumfrequency resonance.For the high-frequency resonance problem,a self-tuning notch filter is designed in this paper.The on-line vibration frequency identification and inertia ratio information are used to set the notch frequency and width of the notch filter to ensure the effectiveness of the high-frequency vibration suppression strategy.Finally,based on the theoretical derivation and simulation analysis,this paper systematically establishes a response strategy for the speed vibration of the AC servo system.The software and hardware platform of the AC servo system based on DSP + FPGA is designed and built,and the remote experiment and comparative analysis are completed on the large inertia motor platform and the drag platform.The experimental results prove the effectiveness of the identification method and suppression strategy in this paper.