Sample processing method and system based on kernel extreme learning machine

A technology of nuclear extreme learning machine and processing method, which is applied in the field of soft sensor modeling and application in industrial production process, which can solve the problems of poor stability of soft sensor model and high operation cost, so as to realize soft sensor, reduce operation cost and reduce manual marking cost effect

Pending Publication Date: 2020-11-24
JIANGNAN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a sample processing method and system based on a nuclear extreme learning machine to solve the technical problems of poor stability of the soft sensor model in the prior art solution and high computational cost when modeling complex chemical process

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
  • Sample processing method and system based on kernel extreme learning machine
  • Sample processing method and system based on kernel extreme learning machine
  • Sample processing method and system based on kernel extreme learning machine

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0130] The method proposed by the present invention is applied to the soft measurement of the sulfur recovery process (SRU, Sulfur Recovery Unit). The SRU collects 2000 sets of data in total, from which 1000 sets of data are selected as a training set, and 1000 sets of data are used as a test set. Initially marked in the training set There are 10 samples, 990 unlabeled samples, and the labeling rate is 1%. The KELM method is used to train the model in the iterative process.

[0131] First, the experimental analysis is carried out on the learning step size of the iterative update process (the number of labeled samples per iteration). As the learning step size num increases, more labeled samples are added to the training set to optimize the KELM model, and the model performance is improved. faster, but the cost of manual labeling also increases. Further, in the case of the same number of labeled samples, the simulation experiment is re-run. The smaller the learning step size in ...

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 provides a sample processing method and system based on a kernel extreme learning machine. The method comprises the following steps: recognizing an unmarked sample and a marked sample ina preprocessing sample set, and determining the unmarked sample set and the marked sample set as processing objects; selecting unmarked samples meeting preset requirements from the unmarked sample set based on a KELM algorithm to form a first sub-sample set; marking samples in the first sub-sample set to obtain a second sub-sample set with marks, and adding the second sub-sample set into the marked sample set to obtain an updated marked sample set; and if the similarity between the marked sample set and the updated marked sample set meets a preset requirement, performing soft measurement based on the updated marked sample set. According to the scheme, the iteration updating speed is high, the stability is high, the method and the system are applied to complex chemical process modeling, the operation cost of active learning can be greatly reduced, the manual marking cost is reduced, and soft measurement of process quality variables is more effectively achieved.

Description

technical field [0001] The invention belongs to the field of soft sensor modeling and application in industrial production process, in particular to a sample processing method and system based on nuclear extreme learning machine. Background technique [0002] In complex industrial processes, some key variables that determine product quality need to be monitored and controlled, but due to the constraints of the site environment and technical conditions, it is difficult to measure these variables online. Soft sensing is a common technique used in industrial processes to solve difficult variable detection. A mathematical model is constructed through a training set to realize real-time estimation of new sample quality variables. Common soft sensor models include support vector regression, artificial neural network, Gaussian process regression and extreme learning machine, etc. [0003] Soft-sensing technology usually requires a large number of labeled samples to complete model ...

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): G06F30/20G06N20/00
CPCG06N20/00G06F30/20
Inventor 熊伟丽代学志
Owner JIANGNAN UNIV
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