Product analysis method and system and device and medium

A product analysis and product technology, applied in the field of intelligent manufacturing and artificial intelligence, can solve the problems of large influence of parameter elimination process threshold setting, difficult association rules, complex influence relationship, etc., to achieve strong flexibility and scalability, auxiliary adjustment and optimization , the effect of strong interpretability

Active Publication Date: 2022-06-03
CHENGDU UNION BIG DATA TECH CO LTD
View PDF42 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] Advantages of the solution: It combines the process and experience of historical parameter analysis in the industrial manufacturing process, and fits the business scenarios of industrial manufacturing; disadvantages of the solution: the processing process relies too much on the manual experience of business experts, and the parameter elimination process is greatly affected by the threshold setting
[0014] Advantages of the scheme: the probability of abnormal types is given through the probability graph model, which has strong explanatory power; disadvantages of the scheme: (1) This method is mainly aimed at the operation and maintenance data, between the production and processing data in industrial manufacturing and the operation and maintenance data (2) When the amount of sample data is far less than the parameter dimension, it is difficult to construct a probability graphical model
Disadvantages of the solution: the extraction of association rules is relatively difficult, especially for the field of root cause detection of industrial product defects. The process is complex, involving many parameters, and the influence relationship between parameters is complicated. It is difficult to effectively extract accurate association rules

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
  • Product analysis method and system and device and medium
  • Product analysis method and system and device and medium
  • Product analysis method and system and device and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0095] where SHAP is an additivity explanatory model inspired by the Shapley value. Shapley value originated from

[0108] Before data analysis, all must be preprocessed. It should be noted that the production and processing data in the industrial manufacturing field,

[0110] (1) Delete and record the process parameters that are highly correlated with each other.

[0112] Therefore, the correlations among all process parameters were first calculated. For two process parameters whose correlation exceeds the preset value

[0114] Preferably, the present invention adopts the Pearson correlation coefficient to calculate the correlation of the numerical process parameters.

[0115] (2) Use multiple independent type selection variables to replace the original single type selection parameter.

[0124]

[0127]

[0146] The SHAP value reflects the direction in which the feature affects the model output ("enhancing the output" or "reducing the output") and the effect of

Embodiment 2

[0150] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0153] In the traditional regression analysis method, the number of bad samples is large enough and the processing status of different processing equipment

[0154] The present invention is described below for the analysis of some unfavorable real data in the production of a certain type of glass panel.

[0157] The failure detection data of the glass panel records the product number of the glass panel, the type of failure, and the severity of the failure.

[0158] At the same time, in the original processing data of the product, only the processing records of the product that retains the quality inspection data are screened.

[0159] Obviously, in other embodiments, the product quality grades can be divided into other multiple grade categories, and can also be repeated

[0161] According to the product numb...

Embodiment 3

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 product analysis method, system, device and medium, and relates to the technical fields of intelligent manufacturing and artificial intelligence, including: using the cleaned and preprocessed data to train to obtain a classification prediction model; using the classification prediction model to obtain each The predicted value of the production process parameter; calculate and obtain the SHAP value of each production process parameter based on the predicted value of each production process parameter, and the present invention can accurately and quickly obtain multiple factors affecting product quality based on the SHAP value of each production process parameter Names of key process links and multiple key production process parameters.

Description

Product analysis method and system, device and medium technical field The present invention relates to intelligent manufacturing and artificial intelligence technical field, particularly, relate to product analysis method and system and devices and media. Background technique The root cause locating method of industrial product defect, be applied to the product quality control and process parameter of industrial manufacturing field optimization task. In the industrial manufacturing process, reducing the defective rate of products is very important for the quality control of enterprises. To accurately and quickly locate the root cause of defects in the production and processing process, that is, to detect and locate the key to affecting product quality Process parameters, and then analyze and adjust key process parameters to improve production processes and improve product quality. The root cause positioning of industrial product defect, prior art is mainly through...

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): G06Q10/06G06Q10/04G06N20/00G06K9/62G06Q10/10
CPCG06Q10/06395G06N20/00G06Q10/103G06Q10/04G06F18/24G06F18/214Y02P90/30
Inventor 不公告发明人
Owner CHENGDU UNION BIG DATA TECH CO LTD
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