Method for supervising and predicting abnormal furnace condition of blast furnace based on tuyere information deep learning

A technology of deep learning and blast furnace conditions, applied in the field of metallurgy, can solve problems such as blindness, excessive subjectivity, and roughness of manual experience operations

Pending Publication Date: 2021-05-18
NORTHEASTERN UNIV
View PDF7 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, manual experience operation has the disadvantages of blindness, roughness, and excessive subjectivity, which will lead to misjudgment; abnormal monitoring of furnace conditions based on expert systems usually requires the establishment of corresponding expert knowledge bases for specific blast furnaces and specific problems

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
  • Method for supervising and predicting abnormal furnace condition of blast furnace based on tuyere information deep learning
  • Method for supervising and predicting abnormal furnace condition of blast furnace based on tuyere information deep learning
  • Method for supervising and predicting abnormal furnace condition of blast furnace based on tuyere information deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0057] The present invention provides a method for supervising and predicting abnormal blast furnace conditions based on deep learning of tuyere information. The process is as follows figure 1 As shown, it specifically includes the following steps:

[0058] Step S1: Collect the video of the blast furnace tuyere swirl area. The original video is continuously extracted at a frequency of 10 minutes per frame. After preprocessing, it is stored in the sample database to form the original tuyere image data set; the preprocessing includes:

[0059] S1A: Aiming at the problem of image blur and noise, the noise type of the image is judged by analyzing the spectrum information and histogram information of the image, and the image of the blast furnace gyration is denoised by using the combination of adaptive median filter and gamma transform. Such as figure 2 shown. The adaptive median filter can not only filter out the salt and pepper noise with high probability, but also better prot...

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 tuyere information deep learning-based blast furnace condition abnormity supervision and prediction method, which comprises the following steps of: extracting a blast furnace tuyere image, and carrying out denoising preprocessing on the tuyere image by adopting a mode of combining adaptive median filtering and gamma transformation; enhancing the picture contrast by combining a gray binarization algorithm and gray up-shift, then carrying out unsupervised classification on a sample image, automatically generating a label, and constructing a tuyere image database; and establishing a TI-ResNet model to extract the characteristics of the blast furnace smelting process, and finally utilizing the trained TI-ResNet model to supervise and predict the real-time or off-line furnace condition of the blast furnace based on the blast furnace tuyere monitoring data obtained in real time or off line. According to the method, closed-loop automatic datamation and automatic tagging can be realized, the established TI-ResNet model is automatically trained based on the updated data set, and online or offline furnace condition abnormity supervision and prediction can be performed by applying the established intelligent algorithm and model.

Description

technical field [0001] The invention belongs to the field of metallurgy technology, and in particular relates to a method for supervising and predicting abnormal blast furnace conditions based on deep learning of tuyere information. Background technique [0002] The blast furnace ironmaking process is very complicated. The blast furnace is airtight and operates with high internal chaos. It has the characteristics of time delay, multi-scale, nonlinearity, and strong system coupling, which makes it difficult to accurately predict and predict the process parameters and smelting performance in real time and online. At this stage, the blast furnace is still regarded as a black box. No matter from theory or numerical analysis, or on-site real-time operation often faces many challenges, the tuyere peephole, as the only window that can observe the smelting process in the furnace, has been widely concerned and studied. The process development direction provides the necessary means f...

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): G06T7/00G06T5/00G06K9/62G06N3/04G06N3/08
CPCG06T7/0004G06T5/002G06T5/007G06N3/08G06T2207/10016G06T2207/20032G06T2207/20081G06T2207/20084G06N3/045G06F18/23G06F18/22
Inventor 李强王子佳王帅郭帅李明明邹宗树
Owner NORTHEASTERN 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