Screening method for near infrared spectrum modeling samples

A technology of near-infrared spectroscopy and near-infrared spectrometer, which is applied in the field of data analysis and mathematical modeling, can solve the problems of high time and economic cost pressure of enterprises, lack of operability, labor, equipment, high consumables, etc., to reduce modeling samples volume effect

Active Publication Date: 2019-08-23
CHINA TOBACCO ZHEJIANG IND
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

This patented technology helps reduce the size of models used during analysis while maintaining high-quality results compared to previous methods that were more accurate but had lower overall efficiency. It also allows for clustered data collection based on spectral patterns instead of random selection or just one at once. Overall, it provides technical benefits such improved precision and effectiveness over existing techniques.

Problems solved by technology

Technological Problem addressed in this patents relates to improving accuracy and efficiency when detecting materials with specific compositions using near infraRed light waves. Current techniques require expensive and complicated equipment and consume large amounts of energy during testing due to long preparation times required before use. Therefore there exists an urgent technical problem where reducing the size of these tests while maintaining good performance without requiring excessively complex procedures.

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
  • Screening method for near infrared spectrum modeling samples
  • Screening method for near infrared spectrum modeling samples
  • Screening method for near infrared spectrum modeling samples

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] In this embodiment, the sample is first marked with a random category (for example, under the three conditions, each sample is randomly marked as one of 1, 2, and 3); The label with the best modeling effect is produced, and the samples with the same label are relatively close; repeat the above process several times, and it can be counted whether any sample is close to other samples; finally, the close samples are combined into one category.

[0039] In modeling, the present invention uses different types of samples to model as much as possible, and in the same type of samples, a small number of samples are selected for modeling; this method can analyze the representativeness between samples without knowing other information, Samples of the same class can be represented by one or several of them; modeling should preferably select samples of different classes to enhance the representativeness of the model.

[0040] The detailed technical plan includes the following steps:

Embodiment 2

[0059] This embodiment adopts the full sample modeling method provided by the prior art, Figure 4 Modeling results for the full-sample modeling approach.

Embodiment 3

[0061] In this embodiment, a random method is used to screen the sample modeling results, Figure 6 Modeling results for screening samples using a random approach.

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

According to the modeling sample screening method based on near infrared spectrum modeling, under the condition that other information is not given, samples are divided into a plurality of classes only through spectrums, and the samples in each class are close. Wherein the approximation definition is that modeling is carried out under a randomly given category, and an approximated sample having arelatively high probability is classified into a category; specifically, through the near infrared spectrum of a modeling sample, a plurality of samples are automatically selected for substance content determination and modeling. Under the condition that mass substance content detection cannot be carried out due to factors such as time and cost, the method reduces the use of modeling samples, anddoes not cause great reduction of modeling precision.

Description

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

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
Owner CHINA TOBACCO ZHEJIANG IND
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