Fuzzy Sliding Mode Trajectory Tracking Control Of Permanent Magnet Spherical Motor Based On Nonlinear Disturbance Observer

Permanent magnet spherical motor(PMSM)is a typical multivariable,strongly coupled,time-varying nonlinear system.With the further development of spherical motor technology and the expansion of application,traditional control algorithms such as PD control and computing torque calculation can’t meet the control requirements of spherical motor applications.The application of various intelligent control strategies makes the nonlinear characteristics of spherical motor control system be solved better.However,the controller designed according to the intelligent algorithm has great conservatism,which makes the controller have very large control amplitude,which is difficult to be realized in practice,which limits the practical application of intelligent control strategy.Aiming at the conservativeness of the existing control strategy,this paper proposes a fuzzy sliding mode control strategy based on the nonlinear disturbance observer.Under the premise of satisfying the requirement of the system control precision,a less conservativeness control method was designed.Using the disturbance observer to observe the external interference,friction,internal parameters and other uncertainties in the system,and to compensate at the control input to achieve the suppression of interference,which reducing the conservative of sliding mode control to a certain extent.And the sliding mode controller can cancel the interference observation disturbance and the interference of the observable part,so as to meet the requirements of system tracking performance.In order to further reduce the chattering of the sliding mode,this paper uses the fuzzy logic to approximate the part,and uses the output gain of the fuzzy controller instead of the switching gain of the sliding mode to ensure the continuity of the control rate.In addition,a position detection method for Halbach array permanent magnet spherical motor is proposed for the problem of spherical motor position detection.The position of the rotor is obtained by solving the three rotation angles.In view of the generation and development of the three-dimensional magnetic field Hall sensor,this paper adopts the position detection method based on the magnetic field of the spherical motor rotor.In order to avoid the rotor magnetic field harmonics and stator coil magnetic field on the rotor position detection,this paper presents a magnetic field sensor configuration.According to the magnetic field model and the rotation transformation theory,the nonlinear relationship between the magnetic induction and the rotation angle is obtained.It is found that the gradient projection method is very accurate in solving the rotation angle.However,if the initial search point is chosen incorrectly,the calculation speed of the algorithm is very slow,and the particle swarm algorithm is computationally fast,but the precision is relatively low.In order to solve the rotation angle quickly and accurately,this paper proposes a position solving algorithm combining the gradient projection method and the particle swarm algorithm,and then the position information of the rotor can be obtained.The simulation results show that the position detection method proposed in this paper can realize high precision detection,and can realize fast position detection.