Neural network inverse system decoupling fuzzy self-tuning proportional–derivative control strategy of a bearingless induction motor

Abstract

In view of the problems that the analytical inverse system decoupling method of bearingless induction motor is sensitive to the change of motor parameters and is greatly affected by unmodeled dynamics, and that of traditional proportional–derivative controller lacking the self-adaptive regulation ability, a neural network inverse system decoupling fuzzy self-tuning proportional–derivative control strategy is proposed for a bearingless induction motor system. First, under the conditions of considering the stator current dynamic of torque winding, and by neural network inverse system method, the bearingless induction motor system is decoupled into four pseudo-linear integral subsystems. Second, the traditional proportional–derivative controller is improved, and the fuzzy control algorithm is used to adjust the parameters of improved proportional–derivative controller adaptively. Thus, a neural network inverse system decoupling fuzzy self-tuning proportional–derivative control system is constructed for a bearingless induction motor. The simulation experimental results show that the proposed control strategy not only can effectively improve the stability, robustness and steady-state control accuracy of bearingless induction motor system, but also can significantly improve the dynamic response speed and the ability to resist the influences of motor parameter change and load disturbance.

Research on the Control System of Bearingless Induction Motor Based on Improved Active Disturbance Rejection Control

“…where ψ 1d and ψ 1q denote the air-gap flux chain on the d-q axis, respectively; I 2sd and I 2sq are the components of the stator current of the suspension winding in the d-q axis, respectively; K M is the Maxwell's force constant; K L is the Lorentz's force constant; F x and F y are the components of the levitation force in the x and y directions for the BIM, respectively; l 1 is the effective length of the core; r is the rotor's outer diameter; and N 1 and N 2 are the effective numbers of turns of the torque and suspension force windings, respectively. The equation of motion for the rotor portion of the BIM may be represented as [7]:…”

“…BIM exhibits the attributes of nonlinearity, multivariable nature, and robust coupling. To gain the best static as well as dynamic performance from a BIM, sophisticated control algorithms need to be researched [7,8]. When there are interference factors in the system or when the degree of nonlinearity grows, the proportion integration differentiation (PID) controller is unable to adapt itself appropriately, leading to a diminished ability to effectively suppress disturbances [9][10][11][12].…”

Adaptive Finite-Time Backstepping Integral Sliding Mode Control of Three-Degree-of-Freedom Stabilized System for Ship Propulsion-Assisted Sail Based on the Inverse System Method

“…The generalized regression neural network (GRNN) is used to identify the vehicle model, which eliminates multivariable coupling characteristics [9]. Within a certain range, the adaptability of data-driven inverse models to uncertainty has been confirmed [10,11]. However, disturbances in complex environments are difficult to predict, which poses challenges to the robustness of the system.…”

Maximum Torque Output Strategy of Bearingless Induction Motor in the Field-Weakening Region

“…Compared with other motors, a bearingless induction motor (BIM) possesses the advantages of simple structure, solid and reliable, uniform air gap, low cost, and low groove pulsation torque. A new electrical transmission scheme is provided for the development of sophisticated transmission equipment, such as high-speed centrifugal pump, flywheel energy storage, and agricultural equipment [6][7][8][9][10].…”