Mechanical equipment multi-working-condition fault prediction method based on attention mechanism

A technology for mechanical equipment and fault prediction, applied in prediction, neural learning methods, instruments, etc., can solve problems such as time-consuming and labor-intensive, failure to find operating conditions, failure to make full use of sensor data time dependence, etc., to improve convergence Effects of Speed ​​and Accuracy

Inactive Publication Date: 2020-12-29
NANJING UNIV OF POSTS & TELECOMM
View PDF1 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Its disadvantages are: the method of obtaining data by this technology is to manually produce features, and the process is extremely complex and difficult, time-consuming and labor-intensive; secondly, this method cannot make full use of the potential time-dependent relationship in sensor data, and cannot find the correct The mechanical production process affects greater operating conditions, thereby providing artificial adjustments to the mechanical production process to facilitate

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
  • Mechanical equipment multi-working-condition fault prediction method based on attention mechanism
  • Mechanical equipment multi-working-condition fault prediction method based on attention mechanism
  • Mechanical equipment multi-working-condition fault prediction method based on attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] Below in conjunction with accompanying drawing, technical scheme of the present invention is described in further detail:

[0050] Such as Figure 1-2 As shown, this embodiment proposes a multi-condition fault prediction method for mechanical equipment based on the attention mechanism, including the following steps:

[0051] Step 1: Take out the original data from the database received and saved on the sensor, and preprocess the original data; first, receive the original data and vectorize it, recorded as T c is the failure cycle of the cth component, c∈{1,2,…,n}; each item is a K+1-dimensional vector {x 1 ,x 2 ,...,x k ,y i}, respectively represent k sensor input eigenvalues ​​and corresponding RUL values; then normalize the data, the calculation formula is That is, the corresponding normalized value of each item in S is Finally, the data is time-divided to form sliding time window data, that is, O={o i |i>=0, ic -L}, where o i =s i 'The window length is L...

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 provides a mechanical equipment multi-working-condition fault prediction method based on an attention mechanism, and the method comprises the following steps: 1), carrying out the preprocessing of data, carrying out the vectorization and normalization of original data, and carrying out the division through a sliding time window; 2) embedding the preprocessed data into a long-short-term memory network layer, obtaining a data long-range dependency relationship through training, and calculating a hidden vector; 3) inputting the hidden vector in the step 2) into the attention layer;4) in the network training process, optimizing the network parameters by using a particle swarm optimization technology; and 5) inputting a result obtained by network training into a full connection layer of the neural network, and performing linear regression calculation by utilizing higher-layer feature representation learned by the full connection layer to obtain RUL corresponding to the periodof the machine for manually adjusting machine operation parameters or operation conditions. According to the invention, the natural language processing technology and the attention mechanism are combined to realize fault prediction under multiple working conditions.

Description

technical field [0001] The invention relates to a fault prediction method, in particular to a multi-working-condition fault prediction method for mechanical equipment, which belongs to the technical field of mechanical equipment. Background technique [0002] With the rapid development of sensor technology and industrial systems, industrial equipment sensor data is increasing rapidly. However, in the process of industrial production and manufacturing, due to complex physical and chemical changes in the production process, mechanical equipment will inevitably experience performance degradation or even failure. Faults in the production process may cause serious industrial production problems and economic losses, so it is particularly important to realize fault prediction and health management (PHM) in industrial production. [0003] Nowadays, in order to realize the failure prediction and health management of industrial production, scholars have proposed the concept of remain...

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): G06Q10/04G06N3/00G06N3/04G06N3/08
CPCG06Q10/04G06N3/049G06N3/08G06N3/006G06N3/044G06N3/045
Inventor 孙雁飞张及棠亓晋许斌
Owner NANJING UNIV OF POSTS & TELECOMM
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