Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Coke oven flame path temperature soft measurement method and system, computer and storage medium

A technology of flue temperature and soft measurement, applied to computer components, calculations, calculation models, etc., can solve problems such as inappropriateness, timing matching cannot be accurate to the same time, etc.

Pending Publication Date: 2020-10-09
CENT SOUTH UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, when using the top temperature of the regenerator to perform soft measurement of the temperature of the coke oven flue, it is necessary to match the temperature of the top of the regenerator with the temperature of the coke oven flue to obtain sample data. Due to the inconsistency of the two variable acquisition methods, the timing matching Therefore, the model trained on this kind of sample data is not sensitive to small error data and has a relatively high tolerance to deviation, and the collected flue temperature itself also has a large Uncertainty
In this case, for the soft sensing of such industrial process parameters, it is not appropriate to build a model on a limited labeled data set with the optimization objective of minimizing the sum of squared errors commonly used

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
  • Coke oven flame path temperature soft measurement method and system, computer and storage medium
  • Coke oven flame path temperature soft measurement method and system, computer and storage medium
  • Coke oven flame path temperature soft measurement method and system, computer and storage medium

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0114] Step1: The obtained marked samples and unmarked samples are 810 groups and 2165 groups respectively, denoted as L |810| and U |2165| , L |810| ={(x 1 ,y 2 ),...,(x 810 ,y 810 )}, U |2165| ={x 811 ,...,x 810+2165}, where missing and abnormal data have been eliminated.

[0115] Step2: Use resampling from L |810| 600 groups of samples are selected from the sample set to form L 1 =L |600| 1 , draw L in the same way 2 =L |600| 2 ;

[0116] Step3: respectively in L |600| 1 and L |600| 2 Build a submodel of Co-εSVR type on R 1 and submodel R 2 , set the number of iterations T = 100;

[0117] Step4: Use the submodel R 1 Calculate U |2165| The output of all variables in , get the pseudo-labeled U |2165| "Dataset, the U |2165| The variables in " are put into the submodel R in turn 2 The training set, retraining to get the sub-model R 2 ” , computing 2165 R 2 ” The ε-insensitive loss value, with the original R 2 Compared with the ε-insensitive loss ...

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 a soft measurement method and system for coke oven flue temperature, a computer and a storage medium. The method comprises the steps that the flame path temperature and the storage top temperature of a coke oven are collected to be sample data, then models f1 and f2 are established through a Co-epsilon SVR modeling method with the minimum epsilon insensitive loss as the optimization target, and a sample set of the model f1 is trained to serve as original sample data, a marked sample set of the training model f2 is a new marked sample formed by taking an error of the model f1 as an output and an original input; finally, the model f1 and the model f2 are combined to construct a prediction target model f. According to the invention, the flame path temperature of the coke oven can be measured more accurately through the constructed prediction target model f, and the reliability of a measurement result is improved.

Description

technical field [0001] The invention belongs to the technical field of temperature detection, and in particular relates to a soft measurement method / system / computer / storage medium of coke oven flue temperature under the background of limited marked samples and noise. Background technique [0002] Coke is one of the main raw materials for ironmaking, metallurgy and chemical industry, that is, the product of coking in coke ovens. The flue temperature of the coke oven refers to the average temperature of each combustion chamber, and the flame temperature of the combustion chamber is about 1600°C. The detection method in the furnace is mainly through the temperature measuring hole on the top of the furnace, and the workers use the infrared thermometer to measure manually at regular intervals. It not only has a large workload and a long measurement period, but also cannot meet the real-time requirements. The existing soft-sensing technology obtains the coke oven flue temperatur...

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): G06K9/62G06N20/00
CPCG06N20/00G06F18/214
Inventor 雷琪郭凤梅
Owner CENT SOUTH UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products