Variable selection method based on partial least square

A technique of partial least squares and variable selection, applied in the field of selection, which can solve problems such as poor forecasting ability and inability to improve forecasting performance.

Inactive Publication Date: 2013-02-13
HARBIN INST OF TECH
View PDF2 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention proposes a variable selection method based on partial least squares in order to solve the problem that the current variable selection method has poor prediction ability and cannot improve the prediction performance.

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
  • Variable selection method based on partial least square
  • Variable selection method based on partial least square
  • Variable selection method based on partial least square

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0014] Specific implementation mode one, combination figure 1 To describe this embodiment in detail, the variable selection method based on partial least squares described in this embodiment includes the following steps:

[0015] Step 1: Preprocess the variables to be analyzed to obtain a standardized variable set;

[0016] Step 2: Calculate the regression vector according to the standardized variable set, each element of the regression vector represents the contribution rate of the corresponding input variable to the output, and the sort order of the contribution rate is the order of decreasing order;

[0017] Step 3: Sort the input variables according to the sort order of the contribution rate of the regression vector to obtain the input variable sequence Xior;

[0018] Step 4: Perform correlation test on all input variables in the input variable sequence Xior;

[0019] Step 5. Select and calculate the input variables to obtain regression coefficients.

specific Embodiment approach 2

[0020] The second embodiment is different from the partial least squares-based variable selection method described in the first embodiment in that the specific process of preprocessing the variables to be analyzed in step one to obtain the standardized variable set is:

[0021] The set of variables to be analyzed constitutes the variable set X, initializes the variable set X, removes the outliers of the variable set X according to the normalization process to obtain a standardized variable set, the standardized variable set is a standardized variable set with a mean value of 0 and a variance of 1. .

[0022] This embodiment ensures the effectiveness of the analysis by preprocessing the variables to be analyzed.

specific Embodiment approach 3

[0023] Embodiment 3 The difference between this embodiment and the variable selection method based on partial least squares described in the first embodiment is that the specific process of calculating the regression vector according to the standardized variable set described in step 2 is:

[0024] The input variables have large redundancy and autocorrelation, which is not conducive to subsequent analysis and processing. According to the leave-one-out cross-check method, determine the number of potential variables in the standardized variable set nlv,

[0025] The number of latent variables nlv is then based on the improved deviation least squares regression method to obtain the regression vector b, each element of which represents the contribution rate of the corresponding input variable to the output.

[0026] The improved deviation least squares regression method described in this embodiment can ensure the stability of digital calculation and increase the calculation speed.

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

A variable selection method based on partial least square aims at resolving the problems of poor prediction capability and incapability to improve prediction performance of the existing variable selection method. The variable selection method based on the partial least square includes preprocessing variables to be analyzed to obtain a standardized variable set and calculating regression vector according to the standardized variable set. Elements of the regression vector indicates contribution rate of corresponding input variables to output, and the order of the contribution rate is reduced gradually. The method further includes ordering the input variables according to the order of the contribution rate of the regression vector to obtain an input variable order Xior, conducting a correlation test on all the input variables in the input variable sequence Xior and conducting selection calculation on the input variables to obtain regression coefficient. The variable selection method is applied to a process control system in the fields of steel, boiler, chemical industry, pharmacy and the like.

Description

Technical field [0001] The invention relates to a selection method, in particular to a variable selection method based on partial least squares. Background technique [0002] Process control systems involve many fields such as steel, boilers, chemicals, pharmaceuticals, etc., and have become an important part of modern industrial production. [0003] Modern process control systems often carry out long-term measurement of many state variables in the production process to obtain a large amount of field monitoring data. How to make reasonable and efficient use of these data to ensure the quality of the final product is one of the main challenges facing industrial process control systems. Especially when the scale of the monitored variables is huge, if all the variables are directly used for analysis, it will cause huge calculation pressure on the system and even affect the control effect of the system. Therefore, it is a reasonable choice to simplify the control system by extracting...

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): G06F19/00
Inventor 尹珅卫作龙王光高会军刘方舟
Owner HARBIN INST OF TECH
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