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

DBN-based multi-dimensional time sequence information driven aeroengine fault diagnosis method

A technology for aero-engine and fault diagnosis, applied in the field of aero-engine

Inactive Publication Date: 2019-09-27
HARBIN INST OF TECH AT WEIHAI
View PDF8 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method to mine the internal characteristics of ACARS data of aero-engines and solve multi-dimensional timing problems for at least some of the above problems

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
  • DBN-based multi-dimensional time sequence information driven aeroengine fault diagnosis method
  • DBN-based multi-dimensional time sequence information driven aeroengine fault diagnosis method
  • DBN-based multi-dimensional time sequence information driven aeroengine fault diagnosis method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] like Figure 1 to Figure 5 As shown, a DBN-based multi-dimensional time-series information-driven aeroengine gas path fault diagnosis method provided by an embodiment of the present invention includes the following steps:

[0045] S1. Collect aero-engine ACARS data, construct an engine sample data set, and divide the training set and test set;

[0046] S2. Normalize the data of the training set and the test set, use the wavelet packet transform method to extract the timing information in the engine sample parameters, and use the dynamic time rounding method to extract the relevant information between the engine sample parameters;

[0047] S3. Vectorize the time sequence information within the parameters and the related information between the parameters, and convert them into one-dimensional vectors as engine sample features;

[0048] S4. Using the engine sample features extracted from the training set to train the fault diagnosis model, the fault diagnosis model first...

Embodiment 2

[0108] Embodiment 2 is basically the same as Embodiment 1, and the similarities will not be described in detail. The difference lies in:

[0109] Due to differences in operating characteristics between different types of engines, in order to minimize such differences, all ACARS data collected in this embodiment come from CFM56-7B series engines. The 9 ACARS parameters collected during aircraft cruise are shown in Table 2. A total of 755 samples were collected, including 707 normal samples and 48 fault samples. The distribution of samples is shown in Table 4. Among them, each sample contains 9 parameters, and each parameter takes 30 sampling points.

[0110] Table 4 Sample distribution

[0111]

[0112] In order to test the diagnostic effect of the method provided by the present invention, three groups of comparisons are set: the first group vectorizes the multidimensional time series, and directly utilizes the DBN_SVM model for fault diagnosis. DBN is a model that connects...

Embodiment 3

[0123] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the computer program, any of the above-mentioned The DBN-based multi-dimensional time series information-driven aeroengine gas path fault diagnosis method described in this item.

[0124] Those of ordinary skill in the art can understand that the implementation of all or part of the processes in the methods of the above embodiments can be completed by the hardware of the electronic device. The electronic device includes a memory, a processor, and stored in the memory and can be executed on the processor As for the running computer program, when the processor executes the computer program, it may include the processes of the embodiments of the above-mentioned methods.

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 relates to a DBN-based multi-dimensional time sequence information driven aeroengine gas path fault diagnosis method which comprises the following steps: collecting aeroengine ACARS data; normalization processing, using the wavelet packet transform method to extract the time sequence information of the parameters and using the dynamic time integration method to extract the correlation information between the parameters; vectorizing the time sequence information in the parameters and the correlation information between the parameters and converting the information into one-dimensional vectors; training a fault diagnosis model, wherein the fault diagnosis model firstly extracts the depth features of the input one-dimensional vectors by using BBN and then diagnoses the fault based on the results of the depth feature extraction by using SVM; using the trained fault diagnosis model to identify the faults of the engine sample features extracted from the test set; performing statistics and evaluation on the fault identification result of the fault diagnosis model; and performing fault identification on the aeroengine ACARS data by using the stored fault diagnosis model so as to obtain the diagnosis result. This method can make full use of the multi-dimensional time sequence information of data and effectively process the high-dimensional features of the data.

Description

technical field [0001] The invention relates to the technical field of aero-engines, in particular to a DBN-based multi-dimensional time series information-driven aero-engine fault diagnosis method. Background technique [0002] As the core component of the aircraft, the aero-engine is very prone to failure due to long-term operation under harsh conditions such as high temperature and high pressure, resulting in the failure of the aircraft to fly normally. Therefore, it is necessary to carry out engine fault diagnosis, thereby improving engine safety and reliability, prolonging engine life and reducing maintenance costs invested by airlines. At present, many aeroengine fault detection methods are based on small deviation data of performance parameters provided by the engine manufacturer (Original Equipment Manufacturer, OEM). Airlines not only need to pay high fees to OEMs to obtain data with small deviations in performance parameters, but once the cooperation between airli...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G01M15/05
CPCG01M15/05
Inventor 钟诗胜张永健付旭云李旭
Owner HARBIN INST OF TECH AT WEIHAI
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