Polyphase motor fault-tolerant control method and system based on multi-neural-network inverse model

Abstract

The invention discloses a polyphase motor fault-tolerant control method and system based on a multi-neural-network inverse model in the field of motor control. The polyphase motor fault-tolerant control method comprises the following steps of: respectively acquiring a normal state inverse model and a fault state inverse module of a motor by adopting a neural network inverse model; selecting a corresponding inverse module and a current converter to be serially connected in front of a hysteresis current comparer by a current detection magnetic-chain observation inverse model switching module under a normal state or a fault state to obtain a constant pseudo-linear system; only considering to convert current of d axis and q axis of a stator, inversely output by the neural network after circuit-opening into four phases of reference current, and only adjusting phase relationships of the rest four phases of current; and when motor model parameters changes because the motor has a fault, approaching to a motor inverse system still accurately without the need of correcting control parameters of additional controllers again. According to the polyphase motor fault-tolerant control method and system disclosed by the invention, the adaptability of a fault-tolerant controller to the winding fault of the motor is improved, and the requirements and the circuit cost of an inverter power device are reduced; and the polyphase motor fault-tolerant control method and system have stronger fault tolerance and robustness.

Description

technical field [0001] The invention relates to the field of motor control, in particular to a multi-phase motor fault-tolerant control method and system, which are used in high-reliability electric transmissions such as electric vehicles and ship propellers. Background technique [0002] At present, the fault-tolerant control methods of multi-phase motors are mainly divided into three categories: the first category is that after a fault occurs, only the amplitude and phase of the remaining phase current are changed to maintain the stator’s rotating magnetic potential, so as to ensure that the output torque of the motor remains unchanged, and On this basis, considering the suppression of torque ripple and the spatial symmetry relationship of the remaining phase windings, the spatially symmetric residual phase fault-tolerant current is obtained, and the fault-tolerant control of multi-phase motors is realized. The second type of fault-tolerant control method is to calculate t...

AI Technical Summary

Problems solved by technology

At present, there is no literature report on the application of the neural network inverse system to the fault-tolerant control of multi-phase motors.

Images