Research On Mechanical Parameter Identification And Oscillation Suppression Technology For Servo Drive System

Servo system,one of the most important control and actuator mechanisms in industrial automation,is widely used in machine tools,textile machinery,robots and other fields.In recent years,the level of research and development of domestic servo technology has been continuously improved,and its market share has grown rapidly.At present,servo drive technology needs to consider more and more mechanical influence,such as elastic connection,backlash,safe operation etc.The new technical difficulties brought from them,is still in the exploration stage and the research on commercial servo technology at home and abroad lacks bright spots.In this paper,the problems of the uncertainties of structure and unknown mechanical parameters,the safety caused by mechanical resonance,the backlash nonlinearity,are organically considered in the servo drive system.The studied innovative and ground-breaking control technologies provides new theoretical and practical exploration for the development of higher performance servo drive products,which plays an important role in achieving catch-up and lead of servo technology.Specific research contents include:Firstly,aiming at the uncertain structure and unknown mechanical parameters of servo drive system,this paper proposes a model structure identification method based on the evaluation function.Based on the least squares algorithm,the identification of both the structure and mechanical parameters is realized.The identification models of one-mass system and two-mass system are established respectively.Mechanical parameters such as inertia,damping and shaft stiffness are identified by the samplings of electromagnetic torque and motor speed.On the basis of assuming the system as a two-mass structure,the evaluation function is established according to the identification result of the motor inertia;and then the system structure can be determined.So that the identification of both the structure and mechanical parameters is realized,and based on which,more accurate model parameters are provided for the subsequent key technologies research.Next,for the two-mass elastic system,the Proportional Integral(PI)control,Model Predictive Control(MPC),and EMPC-PI switching control are sequentially studied to suppress mechanical resonance and realize shaft torque limitation.The parameters of PI controller are designed according to pole configuration method with the same dampings.Due to the limited pole range,the method can not suppress the mechanical resonance phenomenon completely.On this basis,MPC controller is studied and simplified into an explicit model predictive controller(EMPC)for semi-offline and semi-online computation.In order to reduce the computation burden and expand the application range of the algorithm in the servo products,the EMPC-PI switching controller is proposed for the first time.A speed error hysteresis is used as the switching condition.The experimental verification shows that the algorithm can effectively suppress mechanical resonance,realize shaft torque limitation,and guarantee fast dynamic response at the same time.When the transmission device contains components such as gears or ball screws,backlash nonlinearity is further introduced in the drive system.This paper analyzed the mechanism of backlash and proposed an adaptive shaft torque compensation strategy to suppress backlash oscillation in speed loop.Firstly,backlash nonlinearity is modeled as the dead zone,and the mechanism is analyzed by the description function method.It shows that the existence of backlash will reduce the shaft stiffness equivalently,reduce the resonance frequency and aggravate the oscillation amplitude.Then,an adaptive shaft torque compensation algorithm is proposed.By designing the feedback and compensation parameters,the system can be equivalent to a one-mass rigid system,thereby suppressing the resonance caused by the backlash and realizing the shaft torque limiting control.Combined with the load inertia identification technology,the adaptability of the algorithm can be further improved.Then,the research of backlash in speed loop is extended to position loop,and the mechanism of limit cycle oscillation induced in the full closed-loop feedback structure is analyzed and the suppression measure is proposed.The expression of the limit cycle frequency is derived by the describing function method combined with the Nyquist stability criterion.The influence of mechanical parameters on limit cycle characteristics is also analyzed.In order to suppress the limit cycle oscillation,the linear state feedback control method based on the pole configuration technique is adopted.It can compensate the system to a one-mass rigid system equivalently,thereby weakening the influence of backlash nonlinearity and suppressing limit cycle oscillation.Finally,the experimental results show that the proposed method can effectively suppress limit cycle oscillation under different working conditions.