Prediction model for endpoint and product composition of copper-converter smelting based on CNN-GAT algorithm collaboration

Abstract

The endpoint timing of copper-converter blowing directly affects the quality of blister copper, furnace stability, and blowing efficiency. Therefore, enhancing the digitalization and intelligence levels of this process has significant practical importance. This study employed a deep learning algorithm that integrated a convolutional neural network (CNN) and graph attention network (GAT). It utilized CNNs to extract image features from the cooling samples of high-temperature melts. Subsequently, by fusing these image features with various production condition data and constraints through the GAT, a model was constructed to determine the best endpoint and predict the product composition. This model could predict the main elemental content of furnace products and estimate the required blowing time. A dataset comprising 5172 production parameters and images of high-temperature cooling samples from a furnace was established. The model was trained and validated using this dataset, and the results indicated that the model achieved endpoint judgment accuracies of 96.73% and 97.85% for the slag-making and copper-making periods, respectively, on the test set. The average prediction error for the composition across four cycles of copper-converter blowing was as low as 0.705wt%, and the average error in estimating the required blowing time was only 1.94 min. The results of this study provide new methods and insights for the development of intelligent endpoint judgment technologies for copper-converter blowing.

Keywords

endpoint judgment of copper-converter blowing / composition prediction / convolutional neural networ / graph attention network

References

[1] J.J. Fan, J.P. He, Q. Liu, and Z.J. Xie, Application of copper end point on-line monitoring system to judge end point of converter blowing, Nonferrous Met. Extr. Metall., (2018), No. 8, p. 8.

[2] Song HY, Gui WH, Yang CH, Peng XQ. Application of dynamical prediction model based on kernel partial least squares for copper converting. Chin. J. Nonferrous Met., 2007, 17(7): 1201

[3] H.Z. Zhang, B.J. Zhang, and T.Q. Guo, Endpoint determination of converter blowing based on deep fusion of flame and flue gas characteristics, Copper Eng., (2023), No. 6, p. 175.

[4] Jiang F, Liu H, Wang B, Sun XF. Basic oxygen furnace blowing endpoint judgment method based on flame image convolution neural network. Comput. Eng., 2016, 42(10): 277

[5] H.Z. Zhang, Z.A. Xu, G.Y. Liu, T.Q. Guo, and B.J. Zhang, Image intelligent monitoring and auxiliary analysis system of copper converter blowing furnace, Copper Eng., (2024), No. 2, p. 1.

[6] Xu X, Liu DF, Xu JX. Prediction model of copper converter blowing endpoint based on GRNN algorithm. J. Kunming Univ. Sci. Technol. Nat. Sci., 2021, 46(3): 9

[7] Z.K. Hu, X.Q. Peng, J.F. Yao, et al., Research on endpoint forecast of copper smelting converter, Nonferrous Met. Extr. Metall., (2000), No. 6, p. 7.

[8] Mei C, Hu ZK, Peng XQ, Yao JF, Hu J. Converting furnace endpoint prediction model based on neural network and adaptive error compensation. Chin. J. Nonferrous Met., 2000, 10(5): 732

[9] Gao S, Li B, Li ZX, et al. . Multi-task dynamic prediction model and application of copper converter end point based on deep learning. J. Kunming Univ. Sci. Technol. Nat. Sci., 2022, 47(4): 8

[10] Sun XH, Xie YF, Gui WH, Song HY. Research on endpoint prediction model of converting furnace. Appl. Res. Comput., 2007, 24(11): 60

[11] Chen J, Fallah-Mehrjardi A, Specht A, O’Neill HSC. Measurement of minor element distributions in complex copper converting slags using quantitative microanalysis techniques. JOM, 2022, 74(1): 185

[12] R. Zhang, M.Z. Li, L.H. Zhong, C.R. Tong, F.Y. He, and J.D. Huang, Research on prediction model of Fe content in slag during copper converter slag-making period based on image recognition, Nonferrous Met. Extr. Metall., (2022), No. 4, p. 21.

[13] Wang YJ, Zhou C, Li YG, Liu TH. Endpoint prediction of copper converter blowing based on artificial bee colony algorithm and regularized extreme learning machine. 2021 China Automation Congress (CAC), 2021, 2437

[14] Qiu YH, Li MZ, Huang JD, et al. . Judgment model of a copper-converter end point based on a target detection algorithm. JOM, 2024, 76(5): 2563

[15] Zhang RH, Yang J. State of the art in applications of machine learning in steelmaking process modeling. Int. J. Miner. Metall. Mater., 2023, 30(11): 2055

[16] Wang MY, Tang J, Chu MS, Shi Q, Zhang Z. Prediction and optimization of flue pressure in sintering process based on SHAP. Int. J. Miner. Metall. Mater., 2025, 32(2): 346

[17] Xie TY, Zhang F, Li YR, Zhang Q, Wang YW, Shang H. Self-attention-based convolutional parallel network: An efficient multi-input deep learning model for endpoint prediction of high-carbon BOF steelmaking. Metall. Mater. Trans. B, 2024, 55(6): 4271

[18] Y.J. Xia, H.B. Wang, and A.J. Xu, dmPINNs: An integrated data-driven and mechanism-based method for endpoint carbon prediction in BOF, Metals, 14(2024), No. 8, art. No. 926.

[19] Veličković P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y. Graph Attention Networks, 2017, [2024-5-18]

[20] Lu GH, Yu T, Wu YF, Pan ZN, Chen JB, Deng BR. Magat-based method for imputing power loss data in wind farms. Power Sys. Technol., 2024, 48(8): 3391

[21] Gao SH, Cheng MM, Zhao K, Zhang XY, Yang MH, Torr P. Res2Net: A new multi-scale backbone architecture. IEEE Trans. Pattern Mach. Intell., 2021, 43(2): 652

[22] He JC, Xu JC, Zhang L, Zhu JQ. An interpretive constrained linear model for ResNet and MgNet. Neural Netw., 2023, 162: 384

[23] Ma NN, Zhang XY, Sun J. Funnel Activation for Visual Recognition, 2020, 351[2024-6-19], (2020)

[24] Liu C, Meng L, Jiao LB, Huang GH, Wu JW. Smoke recognition under transmission channel based on improved YOLOv5s. Comput. Meas. Control, 2024, 32(12): 172

[25] Cheng QS, Li HL, Wu QB, Ma L, Ngan KN. Parametric Deformable Exponential Linear Units for deep neural networks. Neural Netw., 2020, 125: 281

[26] Li QM, Qiu LY. A snail species identification method based on deep learning in food safety. Math. Biosci. Eng., 2024, 21(3): 3652

[27] R.T. Yang, Y. Fu, Q. Zhang, and L.N. Zhang, GCNGAT: Drug–disease association prediction based on graph convolution neural network and graph attention network, Artif. Intell. Med., 150(2024), art. No. 102805.

[28] J.L. Bian, H. Lu, G.H. Dong, and G.H. Wang, Hierarchical multimodal self-attention-based graph neural network for DTI prediction, Brief. Bioinform., 25(2024), No. 4, art. No. bbae293.

[29] J.K. Hao, J. Liu, E. Pereira, et al., Uncertainty-guided graph attention network for parapneumonic effusion diagnosis, Med. Image Anal., 75(2022), art. No. 102217.

[30] W. Lan, Y. Dong, Q.F. Chen, et al., KGANCDA: Predicting circRNA-disease associations based on knowledge graph attention network, Brief. Bioinform., 23(2022), No. 1, art. No. bbab494.

[31] W.Y. Yang, P.P. Wang, S.P. Xu, et al., Deciphering cell-cell communication at single-cell resolution for spatial transcriptomics with subgraph-based graph attention network, Nat. Commun., 15(2024), No. 1, art. No. 7101.

[32] X. Zhou, J. Yang, Y. Luo, and X. Shen, HNCGAT: A method for predicting plant metabolite-protein interaction using heterogeneous neighbor contrastive graph attention network, Brief. Bioinform., 25(2024), No. 5, art. No. bbae397.

[33] Q. Li, Y.F. Wang, J. Dong, C. Zhang, and K.X. Peng, Multinode knowledge graph assisted distributed fault detection for large-scale industrial processes based on graph attention network and bidirectional LSTMs, Neural Netw., 173(2024), art. No. 106210.

[34] M. Keicher, H. Burwinkel, D. Bani-Harouni, et al., Multimodal graph attention network for COVID-19 outcome prediction, Sci. Rep., 13(2023), No. 1, art. No. 19539.

[35] Scarselli F, Gori M, Tsoi AC, Hagenbuchner M, Monfardini G. The graph neural network model. IEEE Trans. Neural Netw., 2009, 20(1): 61

[36] Hamilton WL, Ying R, Leskovec J. Inductive representation learning on large graphs. NIPS’17: Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017, 1

[37] Zhang PY, Yan YC, Zhang X, et al. . TransGNN: Harnessing the collaborative power of transformers and graph neural networks for recommender systems. Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2024, 1285

RIGHTS & PERMISSIONS

University of Science and Technology Beijing