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

Auxiliary variable simplification method for high-dimensional nonlinear soft sensor model

A nonlinear model and auxiliary variable technology, applied in the field of soft sensor, can solve problems affecting the accuracy and generalization ability of soft sensor, ill-conditioned covariance matrix, and reduced modeling accuracy

Inactive Publication Date: 2013-07-03
重庆缇帅科技有限公司
View PDF2 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are often multiple correlations between the auxiliary variables formed in this way, resulting in the ill-conditioning of the covariance matrix in the modeling, reducing the accuracy of the modeling, destroying the stability of the model, and affecting the accuracy and generalization ability of the soft sensor.

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
  • Auxiliary variable simplification method for high-dimensional nonlinear soft sensor model
  • Auxiliary variable simplification method for high-dimensional nonlinear soft sensor model
  • Auxiliary variable simplification method for high-dimensional nonlinear soft sensor model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] Taking the soft sensor of the conversion rate of industrial product HCN as an example, the reduction of the high-dimensional nonlinear soft sensor model is carried out as follows:

[0058] Step 1: Determine n original auxiliary variables that may be related to the leading variable, collect the values ​​of n original auxiliary variables and leading variables, form a sample set, and the size of the sample set is m, and write the n original auxiliary variable data into a matrix form, the leading variable data is written as a matrix Y=[y 1 ,...,y i ,...y m ] T form, where x i ∈ R n×1 ,y i ∈R, i=1, 2,..., m, the data matrix obtained after normalization is as follows:

[0059]

[0060] Y = [ y 1 - Σ j = ...

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 an auxiliary variable simplification method for a high-dimensional nonlinear soft sensor model, which is characterized by comprising the following steps: I, determining n primary auxiliary variables possibly related to a dominant variable, and acquiring the value data of the n primary auxiliary variables and the dominant variable to form a sample set; II, combining a KICA method and an FNN method and respectively calculating weight values of the n primary auxiliary variables; III, forming a primary auxiliary variable sequence; IV, modeling and determining the optimal auxiliary variable according to the minimum mean square error (MSE); and V, obtaining a simplification model of a soft sensor. On the basis of the optimal modeling effect, the auxiliary variable simplification method can find out an auxiliary variable set containing the least auxiliary variables to carry out modeling on the dominant variable, so as to simplify the auxiliary variables.

Description

technical field [0001] The invention belongs to the technical field of soft sensing, and in particular relates to an auxiliary variable reduction method for high-dimensional nonlinear soft sensing models, which is used to guide the reduction of auxiliary variables in the production process. Background technique [0002] At present, in the fields of industrial process, bioinformatics, environmental protection, food safety, etc., there are a large number of detection problems of nonlinear, complex correlation, and unmeasurable object parameters. The soft sensing technology based on soft computing is precisely under this strong industrial demand. It has flourished and become an effective method to solve this kind of problems, and has broad development prospects. For example, Chinese patent (patent number: 200410017533.7) proposes a soft sensor modeling method based on support vector machine. [0003] In the process of soft measurement, the first problem is the selection of aux...

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): G06F17/50
Inventor 苏盈盈李太福颜克胜姚力忠曾诚
Owner 重庆缇帅科技有限公司
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