Fuel system fault detection method based on self-organizing map neural network

A self-organizing mapping and fuel system technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve the problems of complex fuel system structures and difficult physical models, and achieve high recognition and important engineering practical value. Effect

Active Publication Date: 2019-03-22
SHANGHAI JIAO TONG UNIV +1
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in practice, the structure of the fuel system is complex, and

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
  • Fuel system fault detection method based on self-organizing map neural network
  • Fuel system fault detection method based on self-organizing map neural network
  • Fuel system fault detection method based on self-organizing map neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] Such as figure 1 As shown, in this embodiment, the feature quantity obtained from the pressure waveform diagram of the fuel system is used to establish the SOM network after preprocessing, the heuristic algorithm is used to optimize the network structure, and the genetic algorithm is used to optimize the network parameters; and then through the SOM network The neurons are initialized and the training set is used to iteratively train the network; finally, the test set is used for fault detection and recognition.

[0037] The characteristic quantities include: rising edge width, spray initiation pressure, maximum pressure, waveform amplitude, waveform width, waveform area, submaximum pressure and maximum residual wave width.

[0038] The preprocessing refers to: using the normalization method and the principal component analysis method to operate the feature quantity to obtain a dimensionless scalar quantity.

[0039] The commonly used formula of the normalization method...

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 fault detection method of a fuel system based on a self-organizing mapping neural network, which obtains characteristic quantities from a pressure waveform diagram of the fuel system and is used for establishing a SOM network after being pretreated, adopts a heuristic algorithm to optimize the network structure in turn, and adopts a genetic algorithm to optimize the network parameters; Then the SOM neural network is initialized and trained iteratively by using the training set. Finally, fault detection and identification are carried out by using the test set. According to the pressure waveform diagram of the fuel system, subsequent to that preprocess of the raw data, By determining the network structure and optimizing the network parameters to establish the corresponding self-organizing mapping neural network model, and at the same time improve the network parameters initialization relying on experience to choose defects, be good at dealing with various problems in the fault diagnosis of gas turbine, has a high degree of fault identification, has important engineering practical value.

Description

technical field [0001] The invention relates to a technology in the field of gas turbine operation and maintenance, in particular to a fuel system fault detection method based on a self-organizing map (SOM) neural network. Background technique [0002] With the rapid development of global large-scale equipment technology, its manufacturing services or fault prediction and health management have become the new expansion and development direction of some service manufacturing enterprises. As one of the most advanced power machinery at present, gas turbine is widely used in aviation, shipbuilding, electric power and other industrial fields. As an important part of the gas turbine, the fuel system directly affects the combustion process of the gas turbine, thus determining the performance of the gas turbine. In the prior art, most methods for detecting fuel system failures use oil pressure sensors to collect oil pressure information of high-pressure oil pipes, and establish a p...

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): G06F11/07G06N3/04G06N3/08
CPCG06F11/079G06N3/04G06N3/08
Inventor 郑宇廖江张燚尧方鸣龚海磊夏唐斌
Owner SHANGHAI JIAO TONG UNIV
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