Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Aircraft electromechanical system fault identification method based on random convolutional neural network

A neural network and random convolution technology, applied in biological neural network models, neural architectures, electrical devices in testing and transportation, etc., to achieve the effect of reducing impact and solving reasonable conversion problems

Inactive Publication Date: 2018-12-14
NORTHWESTERN POLYTECHNICAL UNIV
View PDF9 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the deficiencies of the prior art, the present invention provides a method for fault identification of aircraft electromechanical systems based on random convolutional neural networks. Dimension conversion problem; then use the converted two-dimensional time-frequency graph to train the random convolutional neural network, the network uses random pooling and dropout strategies to suppress model overfitting and improve generalization ability, and update network parameters through the momentum stochastic gradient descent algorithm, Complete the construction of the identification model; finally use the random convolutional neural network to complete the identification of aircraft electromechanical system faults

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Aircraft electromechanical system fault identification method based on random convolutional neural network
  • Aircraft electromechanical system fault identification method based on random convolutional neural network
  • Aircraft electromechanical system fault identification method based on random convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0086] This example adopts the bearing failure data from the laboratory of Case Western Reserve University, and the test bearings adopt SKF bearings and NTN equivalent shafts. The faults in this experiment are single-point bearing faults, and the fault diameters are 0.1778mm, 0.3556mm, 0.5334mm, and 0.7112mm. The damage data of 0.1778mm, 0.3556mm, and 0.5334mm use SKF bearings, and other NTN equivalent bearings are used for testing. The fault types are divided into four types: normal, inner ring, outer ring, and rolling element. This example uses the data collected by the accelerometer at the drive end, and its sampling frequency is 12khz.

[0087] The experimental data set consists of 4 sets of driving end data (16×120,000 data points) under different motor speeds and load conditions. According to the type of fault, the degree of damage and the fault location, the fault state can be divided into 16 categories. The training sample set is 1600 samples under 1797rpm rotation s...

Embodiment 2

[0094] The test data in this example comes from the WS-ZHT1 multifunctional rotor test bench in our laboratory, including the multifunctional rotor test system developed by Beijing Pope Co., Ltd., the rotor bench experiment collector, and the SLM-500 eddy current signal conditioner. , Beijing Pope WS-ICP-6 vibration acceleration model conditioner, YG2003 DC speed control power supply, rotor test bench and Advantech 610H industrial computer. The rotor status is divided into 5 types: normal, unbalanced 1, unbalanced 2, rubbing, rubbing and unbalanced. Unbalance 1 and Unbalance 2 are under different speed and load conditions, and the sampling frequency is 10khz. The training sample set is 500 samples, the test sample is 200 samples, each sample contains 1024 sample points, and the 5 types of training samples are as follows: Figure 4 shown.

[0095] Perform short-time Fourier transform on 1024 points of each sample to obtain a time-frequency diagram of 32×32 power spectral dens...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an aircraft electromechanical system fault identification method based on a random convolutional neural network. The method includes the following steps that: firstly, the short-time Fourier transform is adopted to construct a vibration acceleration signal into a two-dimensional time-frequency map with good spatial correlation, and the problem of two-dimensional transformation of a one-dimensional signal is solved; secondly, the transformed two-dimensional time-frequency map is input into the random convolutional neural network, wherein the network adopts a random dropout mechanism to suppress the model overfitting and enhance the generalization ability, and adopts a momentum stochastic gradient descent algorithm to update network parameters and further complete theconstruction of an identification model; and finally, the random convolutional neural network is adopted to identify the fault of an aircraft electromechanical system. The scheme of the invention is good in identification effect and strong in practicability, is simple and practicable, and is suitable for fault identification of an aircraft system.

Description

technical field [0001] The invention belongs to the field of health monitoring of aircraft electromechanical systems, and in particular relates to a fault identification method for aircraft electromechanical systems. Background technique [0002] As an important part of the aircraft, the aircraft electromechanical system's mission is to transfer energy to the aircraft to realize the basic functions of the aircraft. Aircraft electromechanical systems include aircraft environmental control systems, aircraft fuel systems, aircraft hydraulic systems, aircraft power systems, aircraft auxiliary power systems, etc., and each system is composed of a large number of interrelated components. Rolling bearings and rotors are two representative components in the aircraft electromechanical system. Vibration detection for these components, analysis and processing of the acquired vibration signals can realize the identification of aircraft electromechanical system faults. During the operat...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G01R31/00G06N3/04
CPCG01R31/008G06N3/045
Inventor 梁天辰姜洪开王仲生李华星田红波
Owner NORTHWESTERN POLYTECHNICAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com