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

A Multivariate Industrial Process Fault Identification Method

A technology for industrial process and fault identification, which is applied in the direction of instruments, electrical testing/monitoring, control/regulation systems, etc. It can solve the problems of noise information deteriorating the identification effect and inability to make full use of data

Inactive Publication Date: 2016-06-22
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this method has been successfully applied initially, its disadvantages are: (1) Noise suppression is not considered in the calculation of the dissimilarity index, but the noise information will deteriorate the recognition effect; (2) The dissimilarity analysis is directly aimed at the original measurement Variables, unable to take full advantage of higher-order statistics of the data

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
  • A Multivariate Industrial Process Fault Identification Method
  • A Multivariate Industrial Process Fault Identification Method
  • A Multivariate Industrial Process Fault Identification Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0078] Example 1: Take the CSTR system as an example. The CSTR system is a chemical reactor. The solution of material A enters the reactor, and a first-order irreversible chemical reaction occurs to generate material B. The reaction is an exothermic reaction, so it needs to pass through the external clamp. The cooling agent takes away the heat of reaction. In order to ensure the normal operation of the process, it generally has a cascade control system.

[0079] According to the process mechanism, the dynamic mechanism model of the CSTR system is established as follows:

[0080] dc A d t = - k 0 e - E / R T + Q F c...

Embodiment 2

[0177] Embodiment 2: Taking a well-known TE chemical process as an example, the TE chemical process has been widely used in fault diagnosis and process control method research. The TE chemical process consists of five units, with a total of 52 process variables, including 11 operating variables, 19 component variables and 22 continuous measurement variables. The flow chart of the TE chemical process is as follows Image 6 shown. Additional details on TE chemical processes are found in references (J.J. Downs, E.F. Vogel, Aplant-wide industrial process control problem, Computers & Chemical Engineering, 17 (1993) 245-255).

[0178] The simulation data package of TE chemical process can be downloaded from the website http: / / web.mit.edu / braatzgroup / index.html. The data packet contains 1 normal working condition and 21 fault working condition data IDV(1)~IDV(21). The normal working condition data has 500 sampling points and 52 variables. The fault data set is divided into two cat...

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 present invention relates to a kind of multi-variable industrial process fault recognition method, comprises the following steps: (1) collects the normal operating condition data set X of historical database and the known failure pattern data set of K class, calculates the normal operating condition data set The mean mean(X) and standard deviation std(X) are used to standardize the known failure mode data sets to obtain new failure mode data sets. (2) Construct data windows under each failure mode data set, and calculate six statistical variables. (3) Detect process faults, collect real-time fault data S, and perform standardized processing. (4) Perform statistical principal component dissimilarity analysis on the basis of step (3), and calculate the fault identification index FRI between the fault data set to be identified and the known fault mode data set. (5) Sorting the fault identification index FRI to obtain the fault identification result. The present invention is based on statistical principal component dissimilarity analysis. In the dissimilarity analysis, principal component information is extracted, secondary data information is discarded, the influence of noise is suppressed, and high-order statistical information of data can be fully mined.

Description

technical field [0001] The invention belongs to the technical field of industrial process fault identification, and in particular relates to a multivariable industrial process fault identification method based on statistical principal component dissimilarity analysis. Background technique [0002] With the formation of highly integrated and large-scale modern industrial systems, fault diagnosis technology has become a key technology to ensure the safe and stable operation of modern industrial systems. Diagnostic techniques have attracted widespread attention from industrial process engineers and academic researchers. Researchers have proposed a series of data-driven fault diagnosis methods, including Principal Component Analysis (PCA), Independent Component Analysis (ICA), and Partial Least Squares (PLS). Most of the current research on fault diagnosis methods focuses on the problem of fault detection (that is, how to quickly and effectively find process faults), but there ...

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): G05B23/02
CPCG05B23/024
Inventor 邓晓刚徐莹田学民
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com