Unlock instant, AI-driven research and patent intelligence for your innovation.

Aeroengine after-repair performance prediction method based on stacked autoencoder deep learning network

A deep learning network, aero-engine technology, used in forecasting, data processing applications, geometric CAD, etc., can solve problems such as large performance forecast errors

Active Publication Date: 2021-07-06
HARBIN INST OF TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0015] The purpose of the present invention is to solve the shortcomings of the existing technology that the performance prediction error of aero-engine after repair is relatively large, and propose a method for predicting the performance of aero-engine after repair based on a stacked self-encoding deep learning network

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
  • Aeroengine after-repair performance prediction method based on stacked autoencoder deep learning network
  • Aeroengine after-repair performance prediction method based on stacked autoencoder deep learning network
  • Aeroengine after-repair performance prediction method based on stacked autoencoder deep learning network

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0028] Embodiment 1: The method for predicting the post-repair performance of an aero-engine based on a stacked self-encoding deep learning network includes the following steps:

[0029] Step 1: Use the stacked self-encoding deep learning network to extract the features of the aero-engine performance parameter sequence before sending it for repair and the maintenance depth of the aero-engine unit, and obtain the s×c dimension eigenvector matrix P of the performance parameters before sending it for repair. s×c and the s × d dimension of the cell repair depth eigenvector matrix R s×d ;s is the number of engine maintenance cases, c is the eigenvector matrix P of performance parameters before sending for repair s×c The number of columns (s is the number of rows); d is the unit maintenance depth feature vector matrix R s×d the number of columns;

[0030] Step 2: Combine the feature vector of performance parameters before sending for repair and the feature vector of unit maintenan...

specific Embodiment approach 2

[0057] Embodiment 2: The difference between this embodiment and Embodiment 1 is that in the step 1, the sequence of performance parameters of the aero-engine before sending it for repair is specifically:

[0058] where the x s,m For the s-th engine maintenance case (one engine maintenance is a maintenance case), the exhaust temperature margin value of the m-th flight cycle before sending it for repair.

[0059] Other steps and parameters are the same as in the first embodiment.

specific Embodiment approach 3

[0060] Embodiment 3: The difference between this embodiment and Embodiment 1 or 2 is that: in the step 1, the original information matrix of the maintenance depth of the aero-engine unit body is specifically: where the y s,n It represents the quantity of maintenance depth information for the nth unit of the sth engine maintenance case.

[0061] Other steps and parameters are the same as in the first or second embodiment.

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 method for predicting the after-repair performance of an aero-engine based on a stacked self-encoding deep learning network. The invention aims to solve the shortcomings of the prior art that the performance prediction error of the repaired aeroengine is relatively large. The present invention includes: 1. Obtaining the characteristic vector matrix of the performance parameters before sending for repair and the characteristic vector matrix of the maintenance depth of the unit body; 2. Merging the characteristic vector of the performance parameters before sending for repair and the characteristic vector of the maintenance depth of the unit body to obtain the performance characteristic vector after repair ; Three: Using the after-repair performance feature vector and the after-repair performance parameter sequence corresponding to each maintenance case, the BP neural network is used to establish an aero-engine after-repair performance prediction model; The optimization algorithm optimizes c, d, and h to obtain the optimal aeroengine after-repair performance prediction model. The invention is used in the field of engine repair and maintenance.

Description

technical field [0001] The invention relates to the technical field of aero-engine maintenance optimization, in particular to a post-repair performance prediction method for aero-engines. Background technique [0002] Aero-engine is the main power source and bleed air device in civil aircraft and other aircraft, and its working environment is complex and reliability requirements are high. Therefore, the aero-engine needs to be scientifically repaired and maintained throughout its life cycle. Predicting the performance state of an engine after performing a certain depth of maintenance is the basis for maintenance optimization. The post-repair performance of aero-engines is mainly affected by two factors: the performance state before the repair and the maintenance depth. The performance parameters of the engine are time series, and its maintenance depth is a high-dimensional discrete quantity. In the research related to the after-repair performance prediction of aero-engine...

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): G06F30/20G06F30/15G06Q10/04
CPCG06F30/20
Inventor 钟诗胜林琳李臻
Owner HARBIN INST OF TECH