Modeling And Active Control Of The Rotor System Supported By Rolling-Sliding Bearing

With the increasing requirements for product quality,more stringent requirements are put forward for ultra precision machining machines.Improving the spindle rotation accuracy of machine tools is generally achieved by improving the spindle manufacturing and assembly accuracy,which is costly and complex to implement.If the controllability of the spindle center motion of the hydrostatic bearing is utilized,amplitude and phase modulation can be used to match the amplitude and phase of the fundamental frequency error of the rear bearing,in order to reduce the rotation error of the spindle extension end.This method is not only simple to implement,but also can reduce costs.This thesis combined with the axial motion of the front and rear bearings of the rotor system supported by rolling-sliding bearing and the dynamic and static characteristics of the hydrostatic bearings,with the goal of improving the rotational accuracy of the spindle,a comprehensive spindle rotational error model and a nonlinear spindle axial motion model have been established,and the effects of different active control methods on the spindle rotational error under different operating conditions have been studied,It provides a possible theoretical method for the active control of such ultra-precision machine tools.Firstly,a comprehensive spindle rotation error model is established to reveal the formation principle of spindle rotation error.Based on frequency domain identification,a system identification model is established,and a spindle motion model is established based on Reynolds equation and flow conservation equation.Secondly,in order to improve the rotation accuracy of the spindle,the amplitude and phase of the rotation error of the front and rear bearings are controlled to match through amplitude and phase modulation to reduce the rotation error of the spindle.The rotor axial motion at the hydrostatic bearing is obtained by using the established mathematical model of the motion of the hydrostatic bearing and the Euler method.The rotational error data of the rolling bearing at different rotational speeds are measured in a actual system.Through MATLAB simulation analysis,it is found that there is a phase difference between the rotor at the front bearing and the rotor at the rear bearing.The rotor is controlled by the controllable oil film force provided by the hydrostatic bearing embedded with control chamber to control the tangential and normal acceleration of the rotor.The rotor acceleration and deceleration at the front bearing are regulated to match the rotor phase at the rear bearing,reducing the phase difference between the front and rear bearings.After performing phase modulation control on the spindle,the simulation results show that the amplitude of the rotation error end has been reduced by 40.23%.Using PID control algorithm for amplitude modulation control has reduced the eccentricity of the spindle,and the amplitude of the rotation error has been reduced by 94.6%,effectively improving the rotation accuracy of the spindle at the extension end.Then,in order to improve the real-time and accuracy of the control of the spindle axial movement,based on the small perturbation method,the oil film force is locally linearized,and a system state space model is derived.As the position of the spindle axis changes at any time during tracking control,the dynamic characteristic parameters of each state of the spindle are calculated.On this basis,LQR control algorithm based on genetic algorithm optimization and MPC control algorithm are designed and applied to the nonlinear system of hydrostatic bearings.Active control simulations are conducted under different loads,and compared with traditional PID algorithms.The results show that the control accuracy and stability of the two algorithms are better than those of the PID algorithm,greatly reducing the rotation error of the spindle.Under the same control goal,the LQR simulation time length optimized by genetic algorithm is less than the MPC simulation time length,so the LQR optimized by genetic algorithm can consume less resources.Under the verification of three control algorithms,the results verify the feasibility and superiority of the control strategy of tracking the axial motion at the front bearing to the axial motion at the rear bearing.Finally,based on the laboratory rotor system supported by rolling-sliding bearing,a shaft center motion tracking control experiment was conducted under the influence of external loads.Under the condition of applying a constant load in the x and y directions of the main shaft,after applying the control effect,the x and y directions of the main shaft at the front bearing are simultaneously affected by the controllable oil film force provided by the hydrostatic bearing,and the shaft center will track the calculated movement of the shaft center at the rear bearing to match its phase and amplitude.The results show that,The comprehensive amplitude of the spindle extension end has decreased by about 28%,verifying the correctness of the theory that controlling the phase amplitude matching of the front and rear bearings can reduce the rotation error of the spindle.