GPR (gaussian process regression) online soft measurement method with model updating

A technology of model update and soft measurement, which is applied in character and pattern recognition, instruments, computer components, etc., can solve problems that cannot be solved by offline models, and achieve the effect of improving production efficiency

Inactive Publication Date: 2016-10-26
JIANGNAN UNIV
View PDF0 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In the actual industrial process, because the process characteristics and various parameters

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
  • GPR (gaussian process regression) online soft measurement method with model updating
  • GPR (gaussian process regression) online soft measurement method with model updating
  • GPR (gaussian process regression) online soft measurement method with model updating

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0018] Combine below figure 1 As shown, the present invention is further detailed:

[0019] This paper verifies that the method proposed in this paper has higher prediction accuracy through an actual industrial wastewater treatment process simulation.

[0020] Step 1: Collect the input and output data of the sewage treatment process to form a soft sensor modeling database.

[0021] Step 2: Standardize the training samples, and use PCA principal component analysis to obtain the score matrix. The PCA algorithm is:

[0022] If there is a training data set X={x i |x i ∈R m } i=1...n , M is the dimension of the process variable, n is the number of training data. In general, the dimensionality of the sample and the correlation of variables will greatly affect the speed and quality of modeling, while PCA (Principal component analysis) can reduce the dimensionality of high-dimensional data through singular value decomposition. At the same time reduce the correlation between variables. The...

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 GPR (gaussian process regression) online soft measurement method with model updating. Characteristics and various parameters of the actual industrial process are changed with the passage of time, and a soft measurement model built by a traditional off-line method cannot adaptively adjust the model parameters. The invention provides a model updating method integrating local updating and overall updating so as to solve the type of problems. The method provided by the invention comprises the steps of firstly carrying out off-line modeling on a training sample by using a method of gaussian process regression (GRP) so as to acquire a predicted output value and a predicted error; analyzing the predicted error acquired by the off-line model, carrying out overall updating on the GPR model when the error mean is greater than a certain preset threshold, that is, parameters of a covariance matrix and a covariance function of the GPR model are updated simultaneously; otherwise, carrying out local updating on the GPR model, that is, only the covariance matrix of the GPR model is updated; and finally, carrying out compensation on the predicted output of the updated GPR model by using an error gaussian mixture model (EGMM) so as to acquire a final predicted result. The effectiveness of the method provided by the invention is verified through carrying out modeling and simulation on the actual industrial sewage treatment process.

Description

technical field [0001] The invention relates to a GPR online soft-sensing method with model updating, which belongs to the fields of complex industrial process modeling and soft-sensing. Background technique [0002] The increasing development of science and technology makes the industrial production process more and more complex and refined, and at the same time, the requirements for the index monitoring and control of the industrial process are higher. However, it is very difficult to measure these indicators online in real time. Therefore, soft measurement technology came into being and greatly saved the time of manual analysis and the cost of equipment. [0003] At present, the common data-based modeling methods mainly include partial least squares (PLS), artificial neural networks (ANN), least squares support vector machine (LSSVM) and Gaussian process regression (GPR) and so on. Among them, Gaussian process regression is a new machine learning method developed based...

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): G06K9/62
CPCG06F18/214
Inventor 熊伟丽钟怀兵李妍君
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