Prediction and optimization of flue pressure in sintering process based on SHAP

Abstract

Sinter is the core raw material for blast furnaces. Flue pressure, which is an important state parameter, affects sinter quality. In this paper, flue pressure prediction and optimization were studied based on the shapley additive explanation (SHAP) to predict the flue pressure and take targeted adjustment measures. First, the sintering process data were collected and processed. A flue pressure prediction model was then constructed after comparing different feature selection methods and model algorithms using SHAP + extremely randomized trees (ET). The prediction accuracy of the model within the error range of ±0.25 kPa was 92.63%. SHAP analysis was employed to improve the interpretability of the prediction model. The effects of various sintering operation parameters on flue pressure, the relationship between the numerical range of key operation parameters and flue pressure, the effect of operation parameter combinations on flue pressure, and the prediction process of the flue pressure prediction model on a single sample were analyzed. A flue pressure optimization module was also constructed and analyzed when the prediction satisfied the judgment conditions. The operating parameter combination was then pushed. The flue pressure was increased by 5.87% during the verification process, achieving a good optimization effect.

Keywords

sintering process / flue pressure / shapley additive explanation / prediction / optimization

References

[1] Zheng FQ, Guo YF, Xiang JF, Wang S, Yang LZ, Chen F. Improvement of iron ore sintering productivity by redistributing air volume during sintering process. ISIJ Int.. 2022, 62(1): 74

[2] Roshan V, Kumar K, Kumar R, Nageswara Rao GVS. Preparation of iron ore micro-pellets and their effect on sinter bed permeability. Trans. Indian Inst. Met.. 2018, 71(9): 2157

[3] Nyembwe AM, Cromarty RD, Garbers-Craig AM. Effect of concentrate and micropellet additions on iron ore sinter bed permeability. Miner. Process. Extr. Metall.. 2016, 125(3): 178

[4] Zhou H, Lai ZY, Lv LQ, et al. . Improvement in the permeability of sintering beds by drying treatment after granulating sinter raw materials containing concentrates. Adv. Powder Technol.. 2020, 31(8): 3297

[5] Zhou H, Zhou MX, O’dea DP, Ellis BG, Chen JZ, Cheng M. Influence of binder dosage on granule structure and packed bed properties in iron ore sintering process. ISIJ Int.. 2016, 56(11): 1920

[6] Lu LM, Adam M, Edenton A, et al. . Strategies for efficient utilization of CRL magnetite pellet feed in sintering and pelletising. Proceedings 5th Baosteel Biennial Academic Conference. 2013A31

[7] Q. Shi, J. Tang, and M.S. Chu, Evaluation, prediction, and feedback of blast furnace hearth activity based on data-driven analysis and process metallurgy, Steel Res. Int., 95(2024), No. 2, art. No. 2300385.

[8] Shi Q, Tang J, Chu MS. Key issues and progress of industrial big data-based intelligent blast furnace ironmaking technology. Int. J. Miner. Metall. Mater.. 2023, 30(9): 1651

[9] Huang XX, Fan XH, Chen XL, Yang GM, Gan M. Bed permeability state prediction model of sintering process based on data mining technology. ISIJ Int.. 2016, 56(12): 2113

[10] S.H. Wang, H.F. Li, Y.J. Zhang, and Z.S. Zou, A hybrid ensemble model based on ELM and improved AdaBoost.RT algorithm for predicting the iron ore sintering characters, Comput. Intell. Neurosci., 2019(2019), art. No. 4164296.

[11] Liu S, Lyu Q, Liu XJ, Sun YQ, Zhang XS. A prediction system of burn through point based on gradient boosting decision tree and decision rules. ISIJ Int.. 2019, 59(12): 2156

[12] Chen X, Hu J, Wu M, Cao WH. T–S fuzzy logic based modeling and robust control for burning-through point in sintering process. IEEE Trans. Ind. Electron.. 2017, 64(12): 9378

[13] Du S, Wu M, Chen LF, Jin L, Cao WH, Pedrycz W. Operating performance improvement based on prediction and grade assessment for sintering process. IEEE Trans. Cybern.. 2022, 52(10): 10529

[14] S. Du, M. Wu, L.F. Chen, and W. Pedrycz, Prediction model of burn-through point with fuzzy time series for iron ore sintering process, Eng. Appl. Artif. Intell., 102(2021), art. No. 104259.

[15] Goldstein A, Kapelner A, Bleich J, Pitkin E. Peeking inside the black box: Visualizing statistical learning with plots of individual conditional expectation. J. Comput. Graph. Stat.. 2015, 24(1): 44

[16] Kim J, Lee HJ, Lee H. Mining the determinants of review helpfulness: A novel approach using intelligent feature engineering and explainable AI. Data Technol. Appl.. 2023, 57(1): 108

[17] Shapley LS, Shubik M. A method for evaluating the distribution of power in a committee system. Am. Polit. Sci. Rev.. 1954, 48(3): 787

[18] Z. Zhang, J. Tang, M.S. Chu, et al., The amount prediction and optimization of the returned ore generated from sintering process based on SHAP value and ensemble learning, Steel Res. Int., 94(2023), No. 9, art. No. 2300114.

[19] D.W. Jiang, Z.Y. Wang, K.J. Li, and J.L. Zhang, Analysis of blast furnace permeability regulation strategy based on machine learning, Steel Res. Int., 95(2024), No. 3, art. No. 2300590.

[20] Zhang SJ, Lei HG, Zhou ZC, Wang GQ, Qiu B. Fatigue life analysis of high-strength bolts based on machine learning method and SHapley Additive exPlanations (SHAP) approach. Structures. 2023, 51: 275

[21] Takalo-Mattila J, Heiskanen M, Kyllönen V, Määttä L, Bogdanoff A. Explainable steel quality prediction system based on gradient boosting decision trees. IEEE Access. 2022, 10: 68099

[22] Zhou MX, Zhou H, O’dea DP, Ellis BG, Honeyands T, Guo XT. Characterization of granule structure and packed bed properties of iron ore sinter feeds that contain concentrate. ISIJ Int.. 2017, 57(6): 1004

[23] Chabchoub Y, Togbe MU, Boly A, Chiky R. An In-depth study and improvement of isolation forest. IEEE Access. 2022, 10: 10219

[24] H.X. Ma, T. Peng, C. Zhang, C.L. Ji, Y.M. Li, and M.S. Nazir, Developing an evolutionary deep learning framework with random forest feature selection and improved flow direction algorithm for NOx concentration prediction, Eng. Appl. Artif. Intell., 123(2023), art. No. 106367.

[25] Hamilton RI, Papadopoulos PN. Using SHAP values and machine learning to understand trends in the transient stability limit. IEEE Trans. Power Syst.. 2024, 39(1): 1384

[26] M. Jamei, M. Ali, M. Karbasi, et al., Monthly sodium adsorption ratio forecasting in rivers using a dual interpretable glass-box complementary intelligent system: Hybridization of ensemble TVF–EMD–VMD, Boruta–SHAP, and eXplainable GPR, Expert Syst. Appl., 237(2024), art. No. 121512.

[27] Aminifar A, Shokri M, Rabbi F, Pun VKI, Lamo Y. Extremely randomized trees with privacy preservation for distributed structured health data. IEEE Access. 2022, 10: 6010

[28] K. Uçak and G.Ö. Günel, Adaptive stable backstepping controller based on support vector regression for nonlinear systems, Eng. Appl. Artif. Intell., 129(2024), art. No. 107533.

[29] Koley S, Ray T, Mohanty I, Chatterjee S, Shome M. Prediction of electrical resistivity of steel using artificial neural network. Ironmaking Steelmaking. 2019, 46(4): 383

[30] Wang XP, Hu TH, Tang LX. A multiobjective evolutionary nonlinear ensemble learning with evolutionary feature selection for silicon prediction in blast furnace. IEEE Trans. Neural Networks Learn. Syst.. 2022, 33(5): 2080

[31] Wu SL, Zhu J, Fan JX, Zhang GL, Chen SG. Sintering behavior of return fines and their effective utilization method. ISIJ Int.. 2013, 53(9): 1561

[32] X. Zhang, Q. Zhong, C. Liu, et al., Partial substitution of anthracite for coke breeze in iron ore sintering, Sci. Rep., 11(2021), No. 1, art. No. 1540.

RIGHTS & PERMISSIONS

University of Science and Technology Beijing