A Performance Degradation Evaluation Method of Turbine Engine Based on Stacked Denoising Autoencoder

A turbine engine and self-encoder technology, applied in gas turbine engine testing, jet engine testing, etc., can solve the problems of lack of generality, lack of consistent standards, time-consuming training label selection, etc., to improve time correlation, avoid interference, The effect of reducing the involvement of human and expert experience

Active Publication Date: 2020-06-16
HARBIN INST OF TECH
View PDF11 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the traditional turbine engine health factor (HI) curve construction method that requires the use of complex sensor information evaluation criteria to optimize the selection of original sensor information, and extracting degradation features still needs to rely on a large number of expert experience; HI curve construction The training of the model still usually adopts a supervised method, the selection of training labels is time-consuming and there is no consistent standard; it requires the fusion of multiple signal processing methods, and lacks a certain degree of versatility

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
  • A Performance Degradation Evaluation Method of Turbine Engine Based on Stacked Denoising Autoencoder
  • A Performance Degradation Evaluation Method of Turbine Engine Based on Stacked Denoising Autoencoder
  • A Performance Degradation Evaluation Method of Turbine Engine Based on Stacked Denoising Autoencoder

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0035] Specific implementation mode one: combine figure 1 with figure 2 This embodiment will be described. A method for evaluating performance degradation of a turbine engine based on a stacked denoising autoencoder described in this embodiment, the specific steps of the method are:

[0036] Step 1. Use N monitoring units to obtain all sensor monitoring data of the turbine engine; observe and filter out the data x of sensors whose monitoring data changes k , k=1,2,...,Q, Q is the number of sensors whose monitoring data changes;

[0037] Step 2. The data x of each sensor screened out in step 1 k Normalize to the [0,1] interval respectively; the sensor data of a part of the monitoring unit after screening is used as the training set data, and the sensor data of the other part of the monitoring unit is used as the test set data;

[0038] Step 3: Establish a stacked denoising autoencoder network composed of four denoising autoencoders for feature extraction of training set da...

specific Embodiment approach 2

[0048] Specific Embodiment 2: This embodiment further defines the turbine engine performance degradation evaluation method based on stacked denoising autoencoder described in Embodiment 1. In the second step, the data of each sensor is normalized The process is:

[0049] Normalized by x k * =(x k -x k,min ) / (x k,max -x k,min ), where x k * is each sensor data x k Normalized value, x k,max and x k,min Corresponding to the maximum and minimum values ​​of each sensor in each cycle of the turbine engine, respectively.

specific Embodiment approach 3

[0050] Specific implementation mode three: combination image 3 This embodiment will be described. This embodiment further defines the turbine engine performance degradation evaluation method based on the stacked denoising autoencoder described in the second embodiment. The working principle of the first denoising autoencoder is as follows:

[0051] The training set data is used as the input data of the first denoising autoencoder of the stacked denoising autoencoder network, and the first denoising autoencoder passes the random mapping function q D For input data x k Destroy and get the data after adding noise After encoding process f θ1 Generate the output of the hidden layer The output of the hidden layer After the decoding process g θ1' Generate reconstructed data z; input data x k The difference with the reconstructed data z is taken as the reconstruction error L H (x k , z) for training;

[0052] Encoding process f θ1 The specific process is as follows:

...

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 a turbine engine performance degradation evaluation method based on a stacked-denoising auto-encoder, and belongs to the technical field of engine performance degradation evaluation. The turbine engine performance degradation evaluation method solves the problems that traditional multi-sensor data selection needs to rely on complex information evaluation criteria, extraction of degraded features in HI construction depends on a large number of signal processing techniques and expert experience, supervised training method label selection relies on manual participation, and a method is low in universality. Four denoising auto-encoders build stacked-denoising auto-encoders to extract the single node value of input data. Training set data carries out pre-training to thenetwork and uses a BP algorithm to fine-tune parameters. The extracted single node value is considered to be the health factor value at each cycle, and an HI curve of a training set is established. The test set is input to the trained stacked-denoising auto-encoder to obtain the health factor value at each cycle and construction an HI curve. The HI curves of training set and test set are subjectedto smoothing processing respectively, and the HI curves after smoothing processing are evaluated.

Description

technical field [0001] The invention belongs to the technical field of engine performance degradation evaluation, and in particular relates to a method for evaluating turbine engine performance degradation based on a stacked denoising self-encoder. Background technique [0002] As one of the commonly used and important aircraft components, the turbine engine is of great practical significance to ensure its reliable operation for the stable operation of the aircraft and the reduction of maintenance costs. The health factor of a turbine engine, as a characteristic quantity to evaluate its health level, can represent the degradation state or degree of engine health level. It is obtained by mapping different degradation performance variables during operation. The constructed health factor curve characterizes the engine performance degradation process as a monotonous upward or downward trend. By predicting whether the change of the health factor curve reaches the degradation fail...

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 Patents(China)
IPC IPC(8): G01M15/14
CPCG01M15/14
Inventor 赵光权王少军刘小勇刘月峰姜泽东高永成胡聪彭喜元
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap