Research Of Characteristic Model Based All Coefficient Adaptive Control Of Active Magnetic Bearing

The model of magnetic bearing system itself has certain uncertainty.One of the main sources of the model's uncertainty is the coupling between the electromagnetic characteristics of the magnetic field between the electromagnet of the maglev bearing and rotor and the radial degrees of freedom of the magnetic bearing rotor.Another main reason for the uncertainty of the maglev bearing model is that the rotor's mode will be excited when it is disturbed by the specific frequency,which leads to the uncertainty of the magnetic bearing system model.As to the above problems,the classical PID control needs add cross feedback decoupling and phase compensator to solve them well.The robust control based on modern control theory,such as H? control,has a good solution to the above problems,but the design of the controller is more complex.Previous studies have shown that the Characteristic model based allcoefficient adaptive control strategy is effective for solving the nonlinear,strong coupling and stability problem of controlled object when it is excited by unmodeled mode and it's easy to implement.In this paper,the Characteristic model based all-coefficient adaptive control strategy is applied to the position control of the magnetic bearing-rotor system,and the related research is carried out.First,the mathematical models of the test rig adopted in this paper,including rotor-electromagnet,sensor and power amplifier,are analyzed and deduced,and the rotor dynamic analysis under different supporting stiffness is carried out to get the critical speed of each order.The Characteristic model based all-coefficient adaptive theory is introduced,including the theoretical introduction of the characteristic model and the basic principle of the All coefficient adaptive control.Secondly,the Characteristic model based all-coefficient adaptive control method is simulated.The simulation design of the Characteristic model based all-coefficient adaptive controller is realized in simulation software.Besides,a magnetic bearing rotor system simulation platform is built,and on this basis,the anti-interference performance of the system under the step response and the different frequency interference signals is analyzed,simulation study has demonstrates the robustness of the Characteristic model based all-coefficient adaptive controller.The influence of the parameters of the controller on the performance of the system is also simulated and analyzed.Finally,the test platform of the magnetic suspension bearing rotor system is built,and the hardware and software design of the control system is completed.Through frequency sweep,rotor floating and high-speed rotation tests,the performance study of the magnetic bearing system under the control of the Characteristic model based all-coefficient adaptive controller is carried outThe results show that the performance of the magnetic suspension rotor system controlled by the Characteristic model based all-coefficient adaptive controller is more stable than that of the PID+ phase compensation controller,the vibration amplitude of the rotor near the first order critical speed is smaller and the axis orbit is more concentrated.The vibration peak near the critical speed of the first degree of freedom is reduced by 22.4% compared with the PID+ phase compensation controller,and the vibration peak of the second degree of freedom and the third degree of freedom are reduced by 10.3% and 42.1% respectively,and the vibration peak of the fourth degrees of freedom remains basically unchanged.At the same time,in order to ensure the high-speed and smooth operation of the maglev rotor,the magnetic suspension bearing rotor system composed of Characteristic model based all-coefficient adaptive controller has a shorter stability time when it starts,and there is no overhang or overshoot.When the speed of magnetic suspension rotor is close to the first-order bending critical speed,the vibration peak of magnetic suspension rotor under the control of Characteristic model based all-coefficient adaptive control+ phase compensation control is smaller than that of PID+ phase compensation control,and the vibration peak is reduced by 70.7%,49.4%,90.4% and 89.4% in four radial degrees of freedom.