Intelligent diagnosis model of blast furnace wall bonding based on decision tree-deep neural network

Abstract

The condition of the blast furnace is fundamental to ensuring the stability of the entire production process in the iron and steel industry. Traditional methods of monitoring are often prone to human error and are influenced by subjective judgment. As a result, the integration of automated and data-driven technologies has become indispensable in improving both the efficiency and quality of ironmaking. Abnormal furnace states, such as sliding scale, wall bonding, hanging, collapsing and hearth accumulation. To enhance the speed and reliability of blast furnace operations, it is crucial to accurately identify and address the underlying causes of these anomalies. This article introduces an intelligent diagnostic model designed to detect abnormal furnace conditions by combining decision trees and deep neural networks. Diagnostic features are selected based on expert-driven rules using a decision tree, while a deep neural network classification model is employed for diagnosing the furnace conditions. The model is trained using 80% of the available data in chronological sequence, with the remaining 20% reserved as a test set for optimisation. Upon evaluation, the model achieves an accuracy of 81%. A comparative analysis of four diagnostic models is conducted, utilising precision, recall, accuracy and F1-score as metrics, with the proposed model demonstrating superior performance across all criteria. The results demonstrate that the model effectively captures furnace conditions, with diagnostic outcomes closely aligning with actual anomalies, thereby supporting stable furnace operations. After the implementation of the blast furnace abnormal condition intelligent diagnosis system, production efficiency was significantly improved, response times for abnormal condition handling were greatly shortened, and unnecessary downtime was effectively avoided. The fuel ratio was reduced by nearly 2.6 kg·t⁻¹, the utilisation coefficient increased by ∼4%, and the number of significant fluctuations was significantly decreased. This system has provided automation and data support for the ironmaking industry, contributing to the advancement of the sector towards greater intelligence.

Keywords

Blast furnace condition furnace abnormality decision tree deep neural network model diagnosis

Get full access to this article

View all access options for this article.

References

Yang

, et al. Research on expert system for forecasting abnormal furnace conditions in blast furnace. Henan Metall 2011; 19: 5–8. +11. (in Chinese).

Xiao

Zang

Yuan

, et al. Research on abnormal detection of blast furnace condition based on SPC. J Chongqing Technol Bus Univ (Nat Sci Ed) 2023; 40: 59–63. (in Chinese).

Zhao

. Research on BF condition diagnosis method based on improved PCA_SVM. Northeastern University, 2020. (in Chinese).

Zhu

Zhang

Yang

. Anomaly monitoring algorithm for MWPCA blast furnace based on Gaussian mixture model. Chin J Chem Eng 2021; 72: 1539–1548. (in Chinese).

Zhu

Jiang

Gui

, et al. A real-time identification method of blast furnace condition based on res VGGNet//process control committee of Chinese society of automation, Chinese society of automation. In: Proceedings of the 34th China process control conference, 2023, p.1. (in Chinese) [Publisher unknown].

Walke

Cutright

Glowczewski

, et al. A utility decision tree for cyanotoxin treatment with powdered activated carbon. Journal-American Water Works Association, 2023.

Liu

Cui

. A decision tree-based classifier for constructing quantization strategies. Pure Appl Math 2023; 39: 339–349. (in Chinese).

Tuo

Buyun

Wei

, et al. Identification of PCB congeners and their thresholds associated with diabetes using decision tree analysis. Sci Rep 2023; 1: 18322.

Xie

Zhang

Yang

. Improved ID3 decision tree algorithm induced by neighborhood equivalence relations. Comput Appl Res 2022; 39: 102–105. +112. (in Chinese).

10.

Duan

Zhu

Sun

. An improved ID3 algorithm. J Nanyang Norm Univ 2006; 5: 63–65. (in Chinese).

11.

Gajdar

Shu

, et al. Decision tree-based approach for selective symbol execution and fuzzing test. Electr Automation 2023; 45: 98–100. +104. (in Chinese).

12.

Feng

Yang

. Treatment and prevention of wall thickening in 2# blast furnace of Tangshan steel. Hebei Metall 2016; 2: 34–39. (in Chinese).

13.

. Practice of wall thickening treatment in No. 2 blast furnace of Tangshan steel. Henan Metall 2015; 23: 37–40. +56. (in Chinese).

14.

Liu

Yang

Liu

. Analysis and treatment of reasons of thickening of furnace wall in 1# blast furnace (2500 m ∼ 3) of heavy steel. Hebei Metall 2015; 7: 56–58. +79. (in Chinese).

15.

Dong

Liu

, et al. Early warning model of sinter performance based on RFE-DNN. Metall Energy 2023; 42: 3–7. +16. (in Chinese).

16.

Xie

Cai

Huang

, et al. Research on fault diagnosis of ZPW-2000R rail circuit based. On the combination of OC-SVM and DNN. Ind Eng 2023; 26: 154–163. (in Chinese).

17.

Ren

Xiangshuai

, et al. Power line fault diagnosis based on GRU and GWO-KELM[J/OL]. Integr Intell Energy 2024; 46: 54–62. (in Chinese).

18.

Liu

Zhang

. DrSRGAN network based on dropout improvement. Sci Technol Eng 2023; 23: 10015–10022. (in Chinese).

19.

Xin

Taochuan

Yifeng

, et al. RF-DNN-based subtype diagnosis method for breast cancer. Ind Control Comput 2021; 34: 80–81. +85. (in Chinese).

20.

Biao

Bengpeng

Jing

. Diagnosis method of hydraulic pump fault degree based on DNN-CapsNet. Hydraul Pneum 2023; 47: 41–49. (in Chinese).

21.

Zhang

. Fast hardware implementation of remote sensing image classification algorithm based on deep learning. Xi ‘an: Xi'an University of Electronic Science and Technology, 2022. (in Chinese).

22.

Gao

. Research on high-resolution remote sensing image scene classification based on deep learning. Nanjing: Nanjing University of Information Science and Technology, 2023. (in Chinese).

23.

Chai

Xiangli

Ruidong

, et al. Diagnostic value of logistic regression combined with ROC curve in breast cancer. J Henan Higher Med Coll 2023; 35: 485–489. (in Chinese).

24.

Liu

Lyu

Niu

, et al. TE process fault diagnosis and identification based on K-means clustering. Autom Instrum 2020; ; 7: 5–11. (in Chinese).