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

State monitoring method for non-stationary industrial process based on slow feature analysis

An industrial process, non-stationary technology, applied in the direction of program control, comprehensive factory control, comprehensive factory control, etc.

Active Publication Date: 2020-05-29
ZHEJIANG UNIV
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for state monitoring of non-stationary industrial processes based on slow characteristic analysis for the deficiencies of the existing non-stationary production process working condition division and state monitoring technology

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
  • State monitoring method for non-stationary industrial process based on slow feature analysis
  • State monitoring method for non-stationary industrial process based on slow feature analysis
  • State monitoring method for non-stationary industrial process based on slow feature analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0125] Figure 5 The data of normal operation under working condition 2 shown: the results of intermediate statistics under working condition 2 are as follows Figure 6 As shown, the solid line is the intermediate statistics, and the dotted line is the control limit. It can be seen that the sample points are basically below the control limit. The results of intermediate statistics under working condition 3 are as follows Figure 7 As shown in , the SPE statistic is intermittently exceeding the limit, because the current test condition does not match the condition 3, so the model does not match. Further, according to step (8.3), obtain the process status monitoring index, such as Figure 8 As shown, the solid line is the indicator trend, and the dotted line is the threshold. It can be seen that based on T 2 The state monitoring indicators of SPE and SPE statistics are both greater than the threshold, that is, the process is considered to be in normal operating state. The ...

Embodiment 2

[0127] The process working condition is converted from 2 to 3, wherein, the 20th to 120th sampling points are abnormal: the trend of the monitoring index obtained by the method proposed by the present invention is as follows Figure 10 As shown in , the dotted box indicates the 20th to 200th sampling points, it can be seen that the SPE-based monitoring indicators prepared to find process abnormalities. Using PCA method for state monitoring, the results are as follows Figure 11 shown. can be found, T 2 Both the and SPE statistics are below the control limits, and PCA cannot detect process anomalies. The comparison results of Example 2 prove the effectiveness and sensitivity of the proposed method.

[0128] Generally speaking, the state monitoring method based on the present invention can effectively divide the working conditions of the process, and distinguish the normal state and the abnormal state of the process. This is difficult to achieve with traditional non-stationa...

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 state monitoring method for a non-stationary industrial process based on slow feature analysis. According to the method, the slow characteristic analysis and stability testing are adopted to obtain characteristic quantity capable of indicating process working condition information changes, the process non-stability is decomposed into segmentation stability in a characteristic space, and working condition automatic division is carried out on a non-stable process without stable working points; secondly, statistical modeling and state monitoring is performed on each stable working condition by utilizing a stable modeling technology; and finally, the monitoring indexes based on Bayesian reasoning integrate the information of each working condition, and an integrated state monitoring result is provided. According to the method, the understanding of specific process operation characteristics is enhanced, the state monitoring efficiency and the accuracy of an anomalydetection result are improved, the method can be finally applied to an actual industrial production site, and the safe and reliable operation of a non-stationary industrial process and the high-quality pursuit of products are ensured.

Description

technical field [0001] The invention belongs to the field of statistical state monitoring of non-stationary processes, and in particular relates to a situation where there is no working condition indicator variable. Based on slow characteristic analysis, an industrial process with non-stable operation is automatically and orderly divided into different stable operation conditions. , and carry out statistical modeling and state monitoring methods according to the classification results of working conditions. Background technique [0002] In recent years, with the urgent market demand for multi-variety, multi-standard and high-quality products in modern society, industrial production is more dependent on efficient processes that can produce a variety of products, and the safety and reliability of production operations have become the focus of engineers. hot spot. However, fluctuations in process operating conditions will lead to non-stationary characteristics of industrial pr...

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
IPC IPC(8): G05B19/418
CPCG05B19/41885G05B2219/32339Y02P90/02
Inventor 邹筱瑜潘杰
Owner ZHEJIANG UNIV
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