The upper adjustment of the blast furnace, a key aspect of furnace management, plays a vital role in maintaining stable operation. Currently, it largely depends on manual experience, often leading to deviations in burden composition and distribution. To address this, this paper integrates burden structure and charging system design as a multi-objective optimization problem. A FastICA-GELM-WNN model is proposed to predict production indicators. Based on this, an initial parameter setting model using the Multi-objective Grey Wolf Optimizer is developed. A feedback compensation strategy based on an Improved Adaptive Genetic Algorithm is introduced to correct deviations between predicted and target values. Experimental results show that the proposed Model Predictive Control system achieves around 96% prediction accuracy and effectively adapts to production changes. This enables reductions in carbon emissions and energy consumption while ensuring stable iron output, supporting the broader optimization of blast furnace ironmaking.
YuXBShenYS. Numerical study of the influence of burden batch weight on blast furnace performance. Metall Mater Trans B2020; 51: 2079–2094.
2.
BonteLSergeantRDaelmanA, et al.Influence of the coke and burden quality on the productivity of the blast furnace. Metall Res Technol2005; 102: 405–410.
3.
DiZXYanBChengB, et al.The research on the stability of the burden surface shape during adjustments to the blast furnace charging procedure. Metall Mater Trans B2025; 56: 1056–1070.
4.
HongZBZhouHWuJL, et al.Effects of operational parameters on particle movement and distribution at the top of a bell-less blast furnace based on discrete element method. Steel Res Int2021; 92: 2000262.
5.
HolzingerGSchatzlM. Effect of chute start angle and hopper change on burden distribution during the charging process of a bell-less top blast furnace with two parallel hoppers. Powder Technol2022; 395: 669–680.
6.
SuXLZhangSYinYX, et al.Data-driven prediction model for adjusting burden distribution matrix of blast furnace based on improved multilayer extreme learning machine. Soft COmput2018; 22: 3575–3589.
7.
RadhakrishnanVRRamKM. Mathematical model for predictive control of the bell-less top charging system of a blast furnace. J Process Control2001; 11: 565–586.
8.
LiuXChenLGFengHJ, et al.Constructal design of a blast furnace iron-making process based on multi-objective optimization. Energy2016; 109: 137–151.
9.
PorzioGFNastasiGCollaV, et al.Comparison of multi-objective optimization techniques applied to off-gas management within an integrated steelwork. Appl Energy2014; 136: 1085–1097.
10.
HuZHeD. Operation scheduling optimization of gas-steam-power conversion systems in iron and steel enterprises. Appl Therm Eng2022; 206: 118121.
11.
MahantaBKChakrabortiN. Evolutionary data driven modeling and multi objective optimization of noisy data set in blast furnace iron making process. Steel Res Int2018; 89: 1800121.
12.
LiYFWangBZhouZX, et al.Research progress of intelligent ore blending model. Metals (Basel)2023; 13: 379.
13.
ZhangHBHuangQLiFW, et al.A network security situation prediction model based on wavelet neural network with optimized parameters. Digit Commun Netw2016; 2: 139–144.
14.
OjaEYuanZ. The FastICA algorithm revisited: convergence analysis. IEEE Trans Neural Netw2006; 17: 1370–1381.
15.
Maadanpour SafariFEtebariFPourghader ChobarA. Modelling and optimization of a tri-objective transportation-location-routing problem considering route reliability: using MOGWO, MOPSO, MOWCA and NSGA-II. J Optim Ind Eng2021; 14: 83–98.
16.
OjstersekRBrezocnikMBuchmeisterB. Multi-objective optimization of production scheduling with evolutionary computation: a review. Int J Ind Eng Comput2020; 11: 359–376.
17.
ChaabeneWBFlahMNehdiML. Machine learning prediction of mechanical properties of concrete: critical review. Constr Build Mater2020; 260: 119889.
18.
HeidariAImaniDMKhalilzadehM, et al.Green two-echelon closed and open location-routing problem: application of NSGA-II and MOGWO metaheuristic approaches. Environ Dev Sustain2023; 25: 9163–9199.
19.
SrinivasMPatnaikLM. Adaptive probabilities of crossover and mutation in genetic algorithms. IEEE Trans Syst Man Cybern1994; 24: 656–667.
20.
MirjaliliSSaremiSMirjaliliSM, et al.Multi-objective grey wolf optimizer: a novel algorithm for multi-criterion optimization. Expert Syst Appl2016; 47: 106–119.
21.
AzzamMMousaAA. Using genetic algorithm and TOPSIS technique for multiobjective reactive power compensation. Electr Power Syst Res2010; 80: 675–681.
22.
HanZHongLMengJ, et al.Temperature drift modeling and compensation of capacitive accelerometer based on AGA-BP neural network. Measurement (Mahwah NJ)2020; 164: 108019.
23.
De JongKASpearsWMGordonDF. Using genetic algorithms for concept learning. Mach Learn1993; 13: 161–188.
24.
ShenoyPDSrinivasaKGVenugopalKR, et al.Dynamic association rule mining using genetic algorithms. Intell Data Anal2005; 9: 439–453.
25.
DengWZhangXXZhouYQ, et al.An enhanced fast non-dominated solution sorting genetic algorithm for multi-objective problems. Inf Sci (NY)2022; 585: 441–453.
26.
GeLJLiuJHWangB, et al.Improved adaptive gray wolf genetic algorithm for photovoltaic intelligent edge terminal optimal configuration. Comput Electr Eng2021; 95: 107394.
