A Distributed PCA-Based Multi-Condition Fault Monitoring Method

A fault monitoring, multi-working-condition technology, applied in electrical testing/monitoring, testing/monitoring control systems, instruments, etc., can solve problems such as failure to monitor the model and get the desired effect

Active Publication Date: 2020-08-04
JIANGNAN UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The actual industrial production process will switch production conditions according to the demand, so the obtained historical data often contains information of multiple operating states, showing the characteristics of multi-distribution. Under this condition, it is impossible to directly establish a fault monitoring model for the process. ideal effect

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 Distributed PCA-Based Multi-Condition Fault Monitoring Method
  • A Distributed PCA-Based Multi-Condition Fault Monitoring Method
  • A Distributed PCA-Based Multi-Condition Fault Monitoring Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0073] This embodiment provides a multi-working-condition fault monitoring method based on distributed PCA. This embodiment takes a common chemical process—TE process (Tennessee Eastman Process) as an example; the experimental data comes from the TE process, and the TE process 21 faults were monitored; see Figure 4 , the method includes:

[0074] Step 1: Obtain the normal working condition data set, and standardize it to obtain the data set through the LNS method. The LNS method is:

[0075] Assuming that the m-dimensional original process data is , the LNS method uses the local neighborhood mean and standard deviation information of each sample to standardize, so as to normalize each working condition and obtain standardized data with a single distribution. The standardized data is:

[0076]

[0077] in, represents the sample x i Among the a nearest neighbors in X, the distance criterion is determined by the Euclidean distance; represents the sample x i the first ...

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 multi-condition fault monitoring method based on a distributed PCA, and belongs to the field of complex industrial process modeling and fault diagnosis. According to the method, for the problem that data presents a multi-condition characteristic in some complex industrial processes, local neighborhood standardization processing is performed on multi-condition process dataso that the multi-distribution characteristic of the data is eliminated; then PCA decomposition is performed, the most correlated variable construction sub-blocks are selected in extracted differentprincipal component directions, automatic decomposition of the whole process is realized, a distributed fault monitoring model is established, and a corresponding monitoring statistic is obtained; andfinally, fusion of the obtained sub-block monitoring statistics is performed by employing a Bayesian inference method so that online monitoring of faults is realized.

Description

technical field [0001] The invention relates to a multi-working condition fault monitoring method based on distributed PCA, which belongs to the field of complex industrial process modeling and fault diagnosis. Background technique [0002] At present, with the continuous expansion of industrial production scale in modern chemical industry and metallurgy, the complexity of process flow is getting higher and higher. Fault monitoring has become a research hotspot in the field of process control. [0003] Based on this background, multivariate statistical process monitoring (MSPM) has been widely used in the field of process monitoring, among which principal component analysis (PCA) is the most commonly used method in multivariate statistical methods. Dimensionality reduction, eliminating the correlation between variables, and monitoring the process by establishing the statistics of the principal component subspace and residual subspace, and can obtain better monitoring results...

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 熊伟丽顾炳斌马君霞
Owner JIANGNAN 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