Selective integrated real-time learning soft measurement modeling method based on heterogeneous similarity

A modeling method and similarity technology, applied in character and pattern recognition, special data processing applications, instruments, etc., can solve the problem that the similarity criterion is difficult to determine which variables to choose to build a soft sensor model, etc., to improve the prediction performance. Effect

Active Publication Date: 2019-07-23
KUNMING UNIV OF SCI & TECH
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The main technical problem to be solved by the present invention is: the present invention provides a kind of selective integration soft sensor modeling method (EMO-ELWPLS) based on heterogeneous

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
  • Selective integrated real-time learning soft measurement modeling method based on heterogeneous similarity
  • Selective integrated real-time learning soft measurement modeling method based on heterogeneous similarity
  • Selective integrated real-time learning soft measurement modeling method based on heterogeneous similarity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0021] Embodiment 1: as figure 1 Shown: a kind of selective integration soft sensor modeling method based on heterogeneous similarity, concrete implementation steps are as follows:

[0022] (1) Use the distributed control system to collect auxiliary variables in the industrial process as the input variable X of soft sensor modeling, obtain the variable corresponding to the input variable X as the output variable Y through offline experimental analysis, and form a sample set [X, Y] , Where N is the number of samples, M is the input variable dimension, and L is the output variable dimension;

[0023] (2) Divide the sample set [X, Y] into a training set and a validation set, respectively, where the training set and validation set are used for model training and model parameter optimization, and standardize the sample set [X, Y] to get the mean value 0, a new sample set with a variance of 1

[0024] (3) Define a plurality of different similarity functions respectively, and c...

Embodiment 2

[0059] Embodiment 2: The following combines a specific example of penicillin fermentation to illustrate the effectiveness of the instant learning soft sensor method based on heterogeneous similarity integration. The penicillin fermentation process is often used for soft sensor algorithm verification and is a standard industrial process simulation platform. Its production process is a typical multi-period, non-linear batch process. During the cultivation process, two cascade controllers are used to control the flow of acid / alkali and cold / hot water to maintain pH and temperature. Simultaneously, sterile substrate and air are continuously fed into the bioreactor to provide nutrients for cell growth and product formation, and to maintain the oxygen consumption required by the microorganisms. In the reaction process, the penicillin concentration is a very important key indicator. In order to control the product quality and production efficiency, the online prediction of the penici...

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 relates to a selective integrated real-time learning soft measurement method based on heterogeneous similarity, and belongs to the field of process industry soft measurement modeling andapplication. According to the invention, a local weighted partial least squares (LWPLS) algorithm is used as a base learner; a similarity function library is established by defining a plurality of similarity functions, then the similarity functions are selected based on an evolutionary multi-objective optimization algorithm, a base model meeting accuracy and diversity indexes is constructed according to the selected similarity functions, and finally fusion of the instant learning base model is achieved by adopting a Stacking integrated learning strategy. According to the method, the appropriate similarity is selected from the similarity library through an evolutionary multi-objective optimization algorithm to adapt to a complex industrial process, and the prediction precision is effectively improved through an integration strategy.

Description

technical field [0001] The invention relates to the field of soft sensor modeling and application in the process industry, in particular to a selective integration real-time learning soft sensor modeling method based on heterogeneous similarity. Background technique [0002] In modern industrial processes, online detection of controlled parameters is a necessary condition for process control and optimization, and is also a key measure to ensure product quality and safe operation of the production process. However, in a complex industrial site environment, it is usually difficult to obtain these key variables that can directly or indirectly reflect quality information, compared to easily measured data information such as temperature, pressure, and flow. Taking the liquid product concentration as an example, the product concentration information can be obtained through an online analyzer, or through offline laboratory analysis after sampling. These two methods have their own a...

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): G06F17/50G06K9/62
CPCG06F30/20G06F18/214G06F18/25
Inventor 金怀平李建刚
Owner KUNMING UNIV OF SCI & 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