High-speed electrified railway traction motor multi-fault modeling and diagnosis method

Abstract

A high-speed electrified railway traction motor multi-fault modeling and diagnosis method comprises the following steps: (1) dynamically analyzing multiple faults of the traction motor, establishing aleast square model structure for describing multi-fault characteristics near multiple working points of a system, and realizing effective fusion of a mechanism model and a data driving model; (2) constructing a multi-fault separation algorithm based on class imbalance learning to optimize a fault feature set, and obtaining a decoupled fault class set in an unknown fault parameter space of the traction motor; and (3) adopting a multi-model-based fault parameter estimation technology to solve the problems of estimation and convergence of multi-fault time-varying parameters, and designing a two-stage adaptive multi-model diagnosis system to realize quantitative analysis of stability and alarm levels of multi-fault diagnosis. Aiming at the characteristics of time varying, adjacent coupling, inconsistent distribution and the like of multiple faults, the identification precision and the diagnosis efficiency can be effectively improved, and the blindness of traditional fault diagnosis is avoided.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis and fault-tolerant control, and in particular relates to a multi-fault modeling and diagnosis method of a high-speed electrified railway traction motor. Background technique [0002] High-speed trains are an important carrier of electrified railways and represent the latest development direction of intelligent transportation. Traction motor is the core power unit of high-speed trains, and its fault diagnosis and self-healing capabilities are of great significance to ensure the reliable operation of high-speed trains. Traction motors are increasingly experiencing multiple failures due to harsh operating conditions, interconnected physical structures, and cumulative operating hours. These faults have the characteristics of time-varying, adjacent coupling, and inconsistent distribution, and are difficult to be diagnosed by the traction control unit in a timely manner. However, as the faults ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a high-speed electrified railway traction motor multi-fault modeling and diagnosis method to solve the problem that the complex fault alarm level of traction motors is difficult to accurately estimate, which may cause major chain accidents; and based on clustering or classification However, the fault characteristic method cannot effectively deal with the unbalanced distribution in the process of multi-fault separation.

Images