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

Method for extracting product data characteristics in batches by using matrix generalized inverse

A product data, batching technology, applied in database indexing, electronic digital data processing, structured data retrieval and other directions, can solve the problems of waste of data resources, large amount of calculation, analysis of factors affecting model update, etc., to achieve dimensionality reduction and Noise filtering, the effect of controlling the amount of calculation

Pending Publication Date: 2020-12-29
PURPLE MOUNTAIN LAB
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] (1) Some dimensions of data collection are missing. If these missing data are all discarded, data resources will be wasted
[0012] (2) Massive data is used for one-time feature calculation, which requires a large amount of calculation, and the space and time resources of the computer are relatively large.
[0014] (4) In the process of decomposing the stock model and periodically updating the model, the interdependence matrix between the models is highly coupled, and it is not easy to decompose
[0015] (5) It is not easy to judge when the stock model undergoes major changes and needs to be updated; and when the stock model can maintain the original structure
[0016] (6) The dependency relationship and geometric meaning between vector data are not clear, which affects the element analysis of model update, so it is necessary to perform spatial orthogonal decomposition at the data level

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
  • Method for extracting product data characteristics in batches by using matrix generalized inverse
  • Method for extracting product data characteristics in batches by using matrix generalized inverse
  • Method for extracting product data characteristics in batches by using matrix generalized inverse

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0082] In the process of robot electric welding, before all welding, there will be a teaching program recorded according to manual welding. This teaching program has been repeatedly tested by engineers and is the optimal state that can be achieved before formal production. However, after it is actually connected to the production line, under long-term continuous work, the machine and manufacturing materials will gradually change, resulting in a decline in welding quality. The solution is as follows:

[0083] Step 1. Before performing teaching welding, use a data set with a moderate number, generally no more than 10,000, to build a basic model. The parameters of this basic model are not only collected from the data of small-scale experiments, but also from some data that have been actually launched in the past to form the basic model.

[0084] Step 2. In the basic model, for the data from the small-scale experiment part, it is determined as the part of the stock model, and for ...

Embodiment 2

[0091] In the glass production process, the initial parameter design should be carried out according to the laboratory test first, and then enter into batch production. However, the collection of intermediate data in the intermediate process of glass production is very troublesome, and the entire production is in a process of constant change, so a process system that can be continuously adjusted according to the production process is required.

[0092] Step 1: Before entering the batch production of glass, conduct small-scale experiments, and then collect the parameters of the small-scale experiments as the basic parameters of the model; at the same time, collect a part of the original batch production data, which will be used as the updated parameter part of the model. The combination of the two forms the basic model, and the data scale of the basic model is more than 10,000.

[0093] The second step is to perform feature extraction on the data of the small-scale test and the p...

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 method for extracting product data characteristics in batches by using matrix generalized inverse, which comprises the following steps of: performing characteristic extraction and data dimensionality reduction on a corresponding sample on the basis of batch analysis and test results of a certain product production process; data disturbance caused by random factors duringdata acquisition can be filtered out, intrinsic characteristics of the data are used for representing the data, and the result stability is achieved; meanwhile, a batch updating mode is adopted in themethod, and the time complexity and the space complexity of calculation are low.

Description

technical field [0001] The invention relates to the technical field of product information analysis and processing, in particular to a method for extracting product data features in batches by using a matrix generalized inverse. Background technique [0002] In the process of industrial production, the production process is long and there is continuous production; the production process not only needs to be calculated and optimized before starting production, but also needs to be continuously improved and optimized during the production process. However, due to the complexity of the production model, after the small batch production process, if the basic model is updated every time, the calculation amount and calculation error of the design will be particularly large. Therefore, it is necessary to design a model update method that combines local fine-tuning of the model and overall adjustment of the system. [0003] In the calculation process of the model, the number of dat...

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): G06F16/22G06F16/2458
CPCG06F16/22G06F16/2465
Inventor 夏飞鹏祁学豪陈刚
Owner PURPLE MOUNTAIN LAB
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