Unlock instant, AI-driven research and patent intelligence for your innovation.

Industrial process soft measurement modeling method for time dimension expansion extreme learning machine model

A technology of extreme learning machine and time dimension, applied in neural learning methods, biological neural network models, instruments, etc., can solve problems such as sacrificing generalization performance

Inactive Publication Date: 2017-05-31
ZHEJIANG UNIV
View PDF1 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the randomness of the model and the number of hidden layer nodes are often larger than other neural networks, the generalization performance is sacrificed.

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
  • Industrial process soft measurement modeling method for time dimension expansion extreme learning machine model
  • Industrial process soft measurement modeling method for time dimension expansion extreme learning machine model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] The present invention is aimed at the non-linear and dynamic problems of the industrial process, uses the idea of ​​instant learning to obtain the dynamic matrix containing the dynamic performance of samples through the time dimension expansion method to solve the dynamic problems, and combines the extreme learning machine algorithm as a non-linear algorithm to solve the process non-linearity linear problem. The time dimension expansion method can greatly increase the amount of process data, and can solve the problem of poor generalization of extreme learning machines in the case of industrial process samples generally not large enough, and the problem of large sample correlation It will not impose a great burden on the accuracy and speed of the extreme learning machine. This method can not only realize the extraction of model dynamics information by the time dimension expansion method, but also improve the problem of insufficient generalization performance of extreme l...

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 industrial process soft measurement modeling method based on a time dimension expansion extreme learning machine. Aiming to solve the problem that an extreme learning machine method always has insufficient generalization ability in the industrial process, a time dimension weighting method is used for improvement, the advantages of high precision and speed when the extreme learning machine is used for solving the nonlinear problem are utilized, the advantage that a time dimension expansion method can extract the process dynamics is combined together, the problem of insufficient generalization ability of the extreme learning machine and the shortcomings of large dimension and high correlation of the time dimension expansion method are skillfully solved in the combination process, and the high-speed and high-precision industrial process soft measurement modeling method is obtained.

Description

technical field [0001] The invention belongs to the field of industrial process prediction and control, and in particular relates to a soft sensor modeling method of time dimension expansion extreme learning machine. Background technique [0002] In traditional industrial processes, there are many parameters that play a vital role in improving product quality and ensuring safety, such as reaction rate, product component content, etc., but many are often difficult or impossible to measure directly with sensors. The use of on-line analytical instruments that require a large amount of investment for detection often has a large lag and makes the adjustment not timely enough, so that it is difficult to guarantee product quality. We call these variables that play an important role in the industrial process the leading variables, and some other variables that have been measured are called auxiliary variables. The essence of soft measurement is to realize the technical method of pr...

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): G05B13/04G06F17/50G06N3/04G06N3/08
CPCG05B13/042G06N3/08G06F30/20G06N3/047
Inventor 葛志强李雨绅
Owner ZHEJIANG UNIV