Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Fault feature learning and classification method based on 1DCNN and GRU fusion

A 1D-CNN, classification method technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problem of non-image sequence fault feature extraction and classification, etc.

Active Publication Date: 2019-12-10
BEIJING UNIV OF TECH
View PDF3 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this technology cannot meet the fault feature extraction and classification of non-image sequences.

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
  • Fault feature learning and classification method based on 1DCNN and GRU fusion
  • Fault feature learning and classification method based on 1DCNN and GRU fusion
  • Fault feature learning and classification method based on 1DCNN and GRU fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments of the present invention and with reference to the accompanying drawings. The described embodiments are some, but not all, embodiments of the invention.

[0031] In order to solve some of the deficiencies in the prior art, the embodiments of the present invention provide a fusion fault feature learning and classification method, and apply it to chiller fault feature learning and classification. The feature learning and classification method includes the following step:

[0032] Step 101: Obtain equipment operation data set X∈R d×n , where d represents the dimension of the collected data, and n represents the number of samples; the data used in this example comes from the 1043 research project initiated by the American Society of Heating, Refrigerating and Air-Conditioni...

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 water chilling unit fault feature learning and classification method based on 1DCNN and GRU fusion, and solves the problems that an existing method consumes time and is difficult to process mass data. The method comprises the following steps: firstly, preprocessing collected continuous-time fault sequence information under different working conditions to construct a sample set; secondly, carrying out preliminary feature extraction on the sample set by adopting a dimension reduction algorithm; then, creating a 1DCNN-GRU neural network model; during training, enabling fault features of a training sample at the same moment to act as input of a time step of the network model, adopting a cross entropy loss function, model parameters are adjusted in combination with a back propagation BP algorithm, enabling Softmax regression to act as a classification algorithm, conducting iterative updating on the network model parameters, and finishing model training. The methodnot only can extract the dynamic information between the local features of the sequences and the sequences, but also can realize real-time classification diagnosis of faults.

Description

technical field [0001] The invention belongs to the field of condition monitoring and fault diagnosis of water chillers based on a big data environment, and in particular relates to a learning and classification method for chiller fault features based on the fusion of 1DCNN and GRU. Background technique [0002] With the rapid development of computer and sensor technology, modern industrial systems are showing a trend of complexity and integration, and the data reflecting the operating mechanism and status of the system show the characteristics of "big data". As the key equipment of the refrigeration system of the data center, the chiller includes a condenser, a compressor, an evaporator, an expansion valve, a chilled water circulation system and a cooling water circulation system. Its main function is to provide a cold source for the computer room and ensure the normal operation of the data center. . The occurrence of chiller failures will reduce the system performance, ca...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/048G06N3/045G06F18/213G06F18/253
Inventor 王卓峥董英杰
Owner BEIJING UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products