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

Civil aviation engine gas path fault diagnosis method based on CNN and SVM

A fault diagnosis and engine technology, applied in the direction of engine testing, neural learning methods, computer components, etc., can solve the problems of inaccurate judgment of civil aviation engine status, diagnosis of refugee aviation engine, loss of relationship between civil aviation engine parameters, etc., to achieve Airway fault diagnosis of civil aviation engine and the effect of accurate diagnosis of airway fault of civil aviation engine

Inactive Publication Date: 2019-01-01
HARBIN INST OF TECH AT WEIHAI
View PDF5 Cites 25 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The technical problem to be solved by the present invention is to solve the problem that the machine learning fault diagnosis method in the prior art needs to serialize the gas path state data when diagnosing the gas path fault of the civil aviation engine, thereby losing the influence of the mutual relationship between the parameters of the civil aviation engine, resulting in The problem of inaccurate judgment of the status of civil aviation engines and the lack of samples of civil aviation engine failures make it difficult to directly use the CNN method to diagnose civil aviation engine gas path failures

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
  • Civil aviation engine gas path fault diagnosis method based on CNN and SVM
  • Civil aviation engine gas path fault diagnosis method based on CNN and SVM
  • Civil aviation engine gas path fault diagnosis method based on CNN and SVM

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] Such as Figure 4 As shown, a method for diagnosing civil aviation engine gas path faults based on CNN and SVM provided by the embodiment of the present invention includes:

[0067] S1. Obtain the air path state data of the civil aviation engine and construct multiple normal state data matrices and multiple failure symptom data matrices, wherein the normal state data matrix of the civil aviation engine represents the normal samples of the actual normal flight, and the failure symptom data matrix represents the occurrence of a failure Symptoms of failure samples.

[0068] S2. Construct a training set and a test set, wherein the training set is composed of the normal state data matrix of the civil aviation engine, and the test set is composed of the failure symptom data matrix and the normal state data matrix of the civil aviation engine.

[0069] Preferably, in step S1 or S2, the normal state data matrix is ​​labeled as "0", and the fault symptom data matrix of differen...

Embodiment 2

[0079] Embodiment 2 is basically the same as Embodiment 1, and the similarities will not be repeated. The difference is that step S1 includes:

[0080] S1-1. Obtain the failure time t of the civil aviation engine l through the customer notification report (Customer Notification Report, CNR) of the civil aviation engine and the gas path state data of the civil aviation engine in the maintenance report l , that is, the time corresponding to the fault confirmation point, and the fault time t l Air path performance parameters of the first m flight cycles.

[0081] Preferably, it is considered that OEM manufacturers mainly use four gas path performance parameters of DEGT, DFF, DN2 and EGTM to diagnose civil aviation engine gas path faults. In this embodiment, the air path performance parameters include civil aviation engine exhaust gas temperature variation (DEGT), fuel flow variation (DFF), core engine speed variation (DN2) and exhaust gas temperature margin variation (EGTM).

...

Embodiment 3

[0129] Such as Figure 11 As shown, the main technical path of this embodiment is basically the same as that of Embodiment 1, and the main technical path can be divided into three major steps: (1) sample preprocessing, corresponding to steps S1 and S2; (2) training CNN network, establishing civil aviation engine The air path state feature extraction model corresponds to step S3; (3) fault diagnosis corresponds to steps S4, S5 and S6. The similarities will not be repeated, the differences are:

[0130] In this embodiment, the SVM is evaluated by the fault recognition rate in step S5. Preferably, in step S5, the test sample feature set is used to train the SVM to classify various faults, including:

[0131] S5-1. Separate the test sample feature set according to a set ratio to construct a feature training set and a feature test set respectively; the sample ratio of the feature training set and the feature test set can be 1:1.

[0132] S5-2. Preliminarily train the SVM using th...

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 civil aviation engine gas path fault diagnosis method based on CNN and SVM. The method comprises the following steps of acquiring civil aviation engine gas path state data;constructing a training set and a test set; using the training set to train a CNN model; using the trained CNN model to carry out characteristic mining on a sample in the test set to form a test sample characteristic set; using the test sample characteristic set to train the SVM and classifying various faults; and inputting the civil aviation engine gas path state data to be diagnosed into the trained CNN model to obtain a sample characteristic to be tested, and using the SVM to classify to obtain a gas path fault type. In the invention, a convolutional neural network is used to directly process a matrix, the relationship of input parameters and time is considered, and a relationship between the input parameters is considered too. The SVM is used to classify, a limitation that civil aviation engine fault samples are insufficient is well solved, and civil aviation engine gas path fault diagnosis can be effectively and accurately realized.

Description

technical field [0001] The invention relates to the technical field of civil aviation engines, in particular to a method for diagnosing gas path faults of civil aviation engines based on CNN and SVM. Background technique [0002] The gas path state data of civil aviation engines is a typical multi-dimensional time series data. When performing fault diagnosis on the gas path of civil aviation engines, it is necessary to synthesize the changing trends of various monitored gas path performance parameters in order to obtain more accurate fault diagnosis. Through the customer notification report (Customer Notification Report, CNR) fed back to the airline by the civil aviation engine manufacturer (Original Equipment Manufacturer, OEM), it can be found that the OEM manufacturers mainly use the exhaust gas temperature change (Delta Exhaust Gas Temperature, DEGT), The four performance parameters of Exhaust Gas Temperature Margin (EGTM), core engine speed change (Delta Core Speed, DN2...

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): G01M15/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G01M15/00G06N3/045G06F18/2411G06F18/214
Inventor 钟诗胜付旭云张永健付松
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