Design And Implementation Of A Diagnostic System For Generator Rotary Rectifier

The three-stage brushless generator is an important part of modern aeronautic power systems.Since the 21 st century,the aeronautic flight field has developed unprecedentedly,and the safety and reliability of aeronautic power systems have become particularly important.Rotary rectifiers are the core components of three-stage brushless generators,and their fault conditions have become one of the key factors for the normal operation of aeronautic power systems.Therefore,the research on the fault diagnosis technology of rotary rectifier has important research significance and practical value.This paper mainly studies the fault diagnosis platform and diagnosis method of rotary rectifier based on host computer.The specific research contents are as follows:(1)Firstly,through research,read a large number of domestic and foreign three-stage aeronautical generator related data,the development of the field of generator rotary rectifier fault diagnosis at home and abroad is elaborated.Developping PC diagnostic platform for the shortcomings of commonly used rotary rectifier fault diagnosis platform.;(2)Secondly,the commonly used fault feature extraction and diagnosis methods are studied and applied to the host computer platform.Aiming at the problem that the feature extraction algorithm found in the research process is redundant and the platform diagnosis efficiency is low,a feature extraction method based on stack autoencoder and minimum redundancy maximum correlation method is used to improve the running efficiency of autoencoder.For the stochastic initialization problem of extreme learning machine algorithm parameters,this paper uses the fireworks algorithm to optimize it and improve it's diagnostic accuracy;(3)Finally,hardware and software systems including data acquisition and information processing are implemented.The basic verification and debugging of the algorithm is realized by the simulation platform,and the online test of the system is completed by using the laboratory's three-stage generator rotary rectifier fault setting platform.The test results prove that the system can realize online collection and display of data,feature extraction,fault diagnosis and some other tasks.