Fuzzy neural network decoupling controller for bearing-free permanent magnet synchronous motor

Abstract

The invention discloses a fuzzy neural network decoupling controller for a bearing-free permanent magnet synchronous motor. The decoupling controller is composed of a feedback linear controller and a fuzzy neural network inverse controller connected in series. The fuzzy neural network inverse controller is connected with the bearing-free permanent magnet synchronous motor in series. The input of the feedback linear controller is radial displacement specific rate [x*, y*], rotation specific rate omega*, radial displacement x, y and rotation speed omega while the output is eccentric displacement controlled quantity r and rotation speed control quantity omega c. The input of the fuzzy neural network inverse controller is eccentric angle theta, eccentric displacement control quantity r and rotation control quantity omega c while the output is torque winding three-phase current signals [ila, ilb, ilc] and suspension winding three-phase current signal [i2a, i2b, i2c]. The invention directly adopts the rotor eccentric quantity as control signals and directly adopts the eccentric displacement quantity, the rotation speed and the rotation angle as the input quantity, direct control over rotor eccentricity can be realized; a suspension force control structure is simplified; good rotor radio position and motor speed control performance can be obtained; and control efficiency is improved.

Description

technical field [0001] The invention relates to a bearingless permanent magnet synchronous motor in power transmission control equipment, specifically a decoupling controller for the motor, which is used for the control of motor transmission and is widely used in many special applications such as high-speed numerical control machine tools, industrial pharmaceuticals, aerospace engineering, and sealed pumps. field of electrical transmission. Background technique [0002] The bearingless permanent magnet synchronous motor is a complex system with strong coupling, nonlinear multiple inputs and multiple outputs. The traditional PID linear control method uses direct torque control to perform double closed-loop decoupling control on the motor. Coupled control, but the structure of the controller is more complex, and in actual operation, due to the complexity of the motor system, it is difficult to obtain accurate parameters and obtain satisfactory results. The use of nonlinear de...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problems existing in the control technology of the existing bearingless permanent magnet synchronous motor, and propose a fuzzy neural network of the bearingless permanent magnet synchronous motor with simple structure, strong robustness, low sample requirements and fast learning speed Inverse decoupling controller, which combines the respective advantages of fuzzy logic control, neural network control and inverse systems

Images