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

Abnormal detection method for gas circuit of aero-engine based on deep learning and Gaussian distribution

An aero-engine, Gaussian distribution technology, applied in the registration/indication of vehicle operation, instruments, registration/indication, etc., can solve the problems that QAR data is not widely used, the accuracy is low, and the false alarm rate of engine abnormality detection is high. Speed ​​up anomaly detection, short sampling period, and discriminative effect

Active Publication Date: 2017-08-29
HARBIN INST OF TECH
View PDF4 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The purpose of the present invention is to solve the problems that QAR data is not widely used in the existing engine gas path anomaly detection method, the false alarm rate of engine anomaly detection is high and the accuracy is low, and an aeroengine based on deep learning and Gaussian distribution is proposed. Gas path abnormality detection method

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
  • Abnormal detection method for gas circuit of aero-engine based on deep learning and Gaussian distribution
  • Abnormal detection method for gas circuit of aero-engine based on deep learning and Gaussian distribution
  • Abnormal detection method for gas circuit of aero-engine based on deep learning and Gaussian distribution

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0026] Specific implementation mode 1: The specific process of the aero-engine air path anomaly detection method based on deep learning and Gaussian distribution in this embodiment is as follows:

[0027] Step 1. Select a parameter set in the QAR data, and the parameter set includes engine gas path performance parameters and external environment parameters;

[0028] QAR is a quick memory recorder;

[0029] Step 2. Calculate the difference value of the performance parameters of two engines on the same aircraft in the parameter set selected in step 1, and form a new parameter set with the difference value and the external environment parameters as the input of step 3;

[0030] Step 3, select the stacked denoising autoencoder (SDAE) model in the deep learning method to extract data features from the new parameter set in step 2;

[0031] Step 4: Using a Gaussian distribution-based density estimation algorithm to perform anomaly detection on the data features obtained in Step 3 to...

specific Embodiment approach 2

[0032] Embodiment 2: This embodiment differs from Embodiment 1 in that: in the step 1, a parameter set is selected in the QAR data, and the parameter set includes engine air path performance parameters and external environment parameters; the specific process is:

[0033] Select parameter set in QAR data

[0034] Before detecting the abnormality of the gas path of the aero-engine, firstly, according to the structural characteristics and working principle of the aero-engine body, combined with the actual engineering experience, select the parameters that are most closely related to the performance of the gas path of the engine and the most sensitive to the change of the performance of the gas path of the engine to form the engine gas path. Road anomaly detection parameter set.

[0035] The parameter set means:

[0036] S={P 1 ,P 2 ,...P j ,...P k ,P 1 ',P 2 ',...P j ',...P k ',E 1 ,E 2 ,...E q ,...E r}

[0037] In the formula, S is the parameter set selected in th...

specific Embodiment approach 3

[0039] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is: in the described step two, calculate the difference value of two engine performance parameters on the same aircraft in the parameter set selected in step one; The specific process is:

[0040] Δ j =P j '-P j

[0041] In the formula, Δ j is the difference value of the jth gas path performance parameter of the two engines.

[0042] The current research on anomaly detection of air path of aero-engine usually applies anomaly detection algorithm to the operating data of a single engine to identify abnormal points. This has two disadvantages. One is that the gas path performance parameters of the engine will be significantly changed by factors such as operating condition conversion, altitude, temperature, humidity, flight Mach number, etc. This change will lead to the deterioration of the effect of the current anomaly detection method. The second is that each QAR dat...

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 discloses an abnormal detection method for a gas circuit of an aero-engine based on deep learning and Gaussian distribution and relates to an abnormal detection method for a gas circuit of an aero-engine. The invention aims to solve the problems that in an existing abnormal detection method for the gas circuit of the aero-engine, QAR data are not widely applied, the false alarm rate of abnormal detection of the engine is high and the accuracy is low. The method comprises the following steps: I, selecting a parameter set in the QAR data, wherein the data set comprises a performance parameter of the gas circuit of the engine and an external environmental parameter; II, calculating the difference value of the performance parameters of two engines on a same plane in the parameter set, and forming a novel parameter set by the difference value and the external environmental parameter; III, extracting data characteristics of the novel parameter set in the step II by using an accumulating and noise-eliminating automatic coder model in a deep learning method; and IV, performing abnormal detection on the data characteristics obtained in the step III by a density estimation algorithm based on Gaussian distribution to obtain a result. The method disclosed by the invention is used in the technical field of fault diagnosis of the aero-engine.

Description

technical field [0001] The invention relates to an abnormal detection method for an air path of an aero-engine, and belongs to the technical field of aero-engine fault diagnosis. Background technique [0002] As the heart of an aircraft, a healthy aero-engine is undoubtedly crucial to ensure flight safety, reliability and economy. If the abnormalities of the engine cannot be discovered in time and corresponding measures are taken, they may develop into failures, and even lead to flight accidents in severe cases. In order to improve the safety, reliability and economy of aircraft flight, in addition to regular inspection of the engine body, it is also necessary to conduct abnormal detection of engine operating data. Specifically, accurate and timely anomaly detection on engine operating data can allow managers to allocate additional monitoring resources in advance, efficiently schedule preventive maintenance programs, maximize engine time on wing, improve engine reliability,...

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): G07C5/08
CPCG07C5/0808
Inventor 钟诗胜付旭云林琳罗辉
Owner HARBIN INST OF TECH
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