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

The invention discloses an equipment health condition assessment method and device based on a deep neural network

A deep neural network and health status technology, applied in the field of deep neural network, can solve the problems of insufficient model analysis and evaluation ability, difficult to meet the actual needs of equipment health status assessment, etc., and achieve the effect of high equipment health assessment accuracy.

Inactive Publication Date: 2019-05-31
SUZHOU NUCLEAR POWER RES INST +2
View PDF3 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In terms of model training, shallow models are used to demonstrate the complex mapping relationship between signals and health conditions, which leads to obvious inadequacies in the analysis and evaluation capabilities of the model in the face of large device data, and it is difficult to meet the equipment health status assessment under the background of big data. the actual needs of

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
  • The invention discloses an equipment health condition assessment method and device based on a deep neural network
  • The invention discloses an equipment health condition assessment method and device based on a deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] The embodiment of the present invention provides a device health assessment method based on a deep neural network, which is suitable for characterizing complex and changeable characteristics hidden inside device data, see figure 1 , the method can include:

[0035] In step S11 , the device under test is simulated to run under different working conditions, and corresponding vibration frequency domain signals under different working conditions are acquired.

[0036] In this embodiment, the above-mentioned equipment health status evaluation method is to use the vibration frequency domain signals generated during the simulation operation under different working conditions of the equipment (such as various working conditions, various faults, normal operation, etc.). The domain signal can better characterize the complex and changeable characteristics hidden in the device data, and can be more prepared to identify the health status of the device when faced with complex monitor...

Embodiment 2

[0055] An embodiment of the present invention provides a device health assessment device based on a deep neural network, which implements the method described in Embodiment 1, see figure 2 , the device may include: an acquisition module 100 , a training module 200 , an adjustment module 300 , and a processing module 400 .

[0056] The acquisition module 100 is used to simulate the operation of the device under test under different working conditions, and acquire corresponding vibration frequency domain signals under different working conditions.

[0057] In this embodiment, the above-mentioned equipment health status evaluation method is to use the vibration frequency domain signals generated during the simulation operation under different working conditions of the equipment (such as various working conditions, various faults, normal operation, etc.). The domain signal can better characterize the complex and changeable characteristics hidden in the device data, and can be mor...

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 equipment health condition assessment method and device based on a deep neural network. The method comprises the following steps: simulating operation of a to-be-tested device in different working states, and obtaining corresponding vibration frequency domain signals in different working states; Randomly selecting a preset number of vibration frequency domain signals assample data, and training a preset deep neural network by adopting a DAE algorithm; And evaluating the health state of the to-be-tested equipment by adopting the trained deep neural network. Method provided by the invention, the characteristics of equipment big data and the advantages of the deep neural network are combined; equipment big data fault feature self-adaptive extraction and equipment health condition identification can be completed at the same time; The method has the advantages that the fault information contained in the health condition signal frequency spectrum can be adaptivelyextracted, high equipment health assessment precision is achieved, complex and changeable characteristics hidden in equipment data can be better represented, and the health condition of equipment canbe more accurately identified when complex monitoring and diagnosis tasks are carried out.

Description

technical field [0001] The present invention relates to the technical field of deep neural networks, in particular to a method and device for evaluating equipment health status based on deep neural networks. Background technique [0002] Equipment health status assessment is very important in the field of equipment maintenance and repair, especially the development of Internet technology, which makes the use of Internet technology to intelligently judge the health status of equipment has become an increasingly important technical means. [0003] The traditional equipment health assessment method is based on "feature extraction of signal processing + machine learning model". The traditional method has the following shortcomings: [0004] In terms of feature extraction, it is necessary to master a large number of signal processing technologies combined with rich engineering practice experience to extract fault features, and treat the two links of feature extraction and health ...

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): G06N3/04G06N3/08G01M99/00
Inventor 崔妍黄立军陈世均陈捷飞江虹张圣韩阳
Owner SUZHOU NUCLEAR POWER RES INST