A Fuzzy Neural Network Control System for Permanent Magnet Synchronous Motor Used in Electric Vehicles

Abstract

The present invention discloses a permanent magnet synchronous motor fuzzy neural network control system for an electric car, relates to an electrical transmission and control technology field, and provides a speed controller based on the fuzzy mathematics and neural network theory and a novel sliding-mode observer based on a tracking differentiator. The system comprises a fuzzy neural network control unit, a sensorless unit, a flux linkage and current calculation unit, a dual- current-loop vector control unit and a control object unit, can realize parameter autotuning of the permanent magnet synchronous motor and high-precision speed regulation in the condition without a mechanical speed sensor, can be applied on an electric car taking the permanent magnet synchronous motor as a power device, and is simple in structure and reliable in operation. Compared to a traditional PID speed controller and a sliding-mode observer, the permanent magnet synchronous motor fuzzy neural network control system for an electric car is higher in tracking precision, stronger in robustness and smaller in counter electromotive force buffeting; and when parameter perturbation of the controller or load disturbance, the permanent magnet synchronous motor fuzzy neural network control system for the electric car also can perform online regulation of parameters of the controller and accurately estimate the position and the speed of a motor rotor.

Description

technical field [0001] The invention belongs to the technical field of electric transmission and control, in particular to a fuzzy neural network control system for a permanent magnet synchronous motor used in an electric vehicle. Background technique [0002] The permanent magnet synchronous motor is a multi-variable, strongly coupled nonlinear dynamic system with simple structure, small size, light weight, low loss, low moment of inertia, high power density, high power factor, high efficiency and other physical characteristics, and it is easy to achieve high-speed operation , braking, forward and reverse switching, wide range of speed regulation, good dynamic response performance, and are widely used in the field of electric vehicles. During its operation, there is interference from external disturbances of sudden load changes; with the change of motor operating status, motor parameters will also change to a certain extent; there are also problems such as limited detection...

AI Technical Summary

Problems solved by technology

In addition, the parameters of the traditional Sigmoidal activation function of the BP network are fixed, and its mapping range, slope and position are immutable.

There is an irreconcilable contradiction between the learning ability of the traditional BP neural network and the complexity of the network, which restricts its nonlinear mapping ability, and the learning speed and accuracy are not ideal

Therefore, in the past ten years, researchers have done in-depth research on the shortcomings of the traditional BP neural network algorithm, and proposed many improved algorithms, such as the method of using momentum to speed up offline training, and the method of normalized weight update technology. , fast propagation algorithm, extended Kalman filter method, second-order optimization and optimal filter method, etc., although the network performance has been improved, but in the network training process, the activation function can only adjust the weight, and cannot automatically find the optimal function body

Therefore, the neural network is easy to fall into a local minimum point, the convergence speed is slow, and the generalization ability is weak.

[0004] The accurate acquisition of the rotor position and speed is the key to the stable and fast operation of the motor. At present, the sampling of rotor information in most motor control systems relies on mechanical position sensors such as resolvers and photoelectric encoders. The cost is high, and the occupied area and the inertia of the bearing Big

Images