Research On Fault Diagnosis Technology Of Aircraft Generator Rotating Rectifier Based On Deep Learning Network

The important electrical equipment brushless AC synchronous generator(BASG)is widely used in many fileds.The BASG is often used as the aircraft generator in aircraft AC power system in aviation field,and it is also the important part of the main power supply of the aircraft.In recent years,aviation accidents and even casualties caused by faulty aircraft generator happen occasionally,the safety and reliability of the aircraft generator have become increasingly important.Rotating rectifier is a crucial but fragile part in aircraft generator,so the diagnosis of the rotating rectifier is of great significance which has important practical and practical value.This paper mainly studies the fault diagnosis method of rotating rectifier based on deep learning network,the main contents are as follows:(1)First,this paper studies on the fault detection and diagnosis methods of rotating rectifier of aircraft generator.The deep learning theory is introduced into the field of fault diagnosis of three-phase rotary generators of aviation,and is also compared with the traditional fault diagnosis method.A fault diagnosis method of rotating rectifier based on deep belief network which can adaptively extract fault features and classify faults is researched.(2)Secondly,in order to verify the influence of the starting and ending points of the fault signal's phase angle on the diagnosis effect,this paper studies a fault diagnosis method based on stack auto-encoder and minimum distance classification method,the experimental verification of the proposed method are shown in the paper.(3)Finally,a fault detection method of rotating rectifier based on stack auto-encoder and support vector data description algorithm is researched in the paper,the method realizes the discrimination of aircraft generator rotating rectifier diode,when only have datas under normal working condition as the train samples.