Coke Reactivity Research Articles

A compact furnace for in-situ studies on high-temperature reactivity of coke using synchrotron radiation SAXS

The study of coke reactivity under high temperature conditions is crucial for understanding its behavior in industrial processes. This study presents a novel compact heating furnace, developed for in-situ synchrotron small-angle X-ray scattering (SAXS) studies on the high-temperature reactivity of coke. The furnace achieves temperatures up to 1400∘C at a heating rate of 12∘C/min and enables steam introduction, simulating industrial conditions. Combination of the furnace with the in-situ SAXS setup at synchrotron radiation facilities facilitates real-time monitoring of structural changes at nanoscale in coke under various temperature and atmospheric conditions. This advancement offers a robust experimental platform for studying coke reactivity, with potential benefits for optimizing industrial processes and reducing environmental impact.

The solution-loss reaction of cokes with different reactivity and lump ores during reductive coupling

ABSTRACT Five cokes with different reactivities and one mixed coke were coupled with lump ores to explore the changes in deterioration behavior under a CO2-to-N2 volume ratio of 30:70. Macro characteristics and microstructures of cokes with different reactivities were analyzed. From a macro perspective, cokes’ solution-loss reaction changed significantly when coke reactivity (CRI) ≈ 35%. Cokes mainly adhered to the uniformly dissolving model when CRI < 35%. Additionally, the average particle size of cokes exhibited negligible changes, merely decreasing from 8.34 to 8.19 mm; however, pore walls suffered severe damage. Cokes mainly adhered to the nuclear-unreacted model CRI > 35%. Besides, the average particle size of cokes has significantly decreased to 7.51 mm, and pore wall damage predominantly occurred on the surface. The conversion of the reaction model increased the difficulty in separating cokes and ores. However, mixed cokes were excellent in slowing down pore wall damage and cokes’ solution loss, which reduced costs in actual production. The order degree of the carbon structure significantly increased at a micro level following the coupling reaction, which led to an enhanced anti-dissolution ability compared to unreacted cokes. Besides, the fringe separation distance of the carbon structure did not change significantly with increased reactivity. However, the order degree gradually decreased accompanied by the decreased anti-dissolution ability.

Improved activity and coking-resistance of Fe2O3/ZrO2 catalysts by hydrothermal synthesis in propane dehydrogenation with CO2: Experimental, DFT calculations, and deactivation modeling

In this work, two ZrO2-supported Fe2O3 catalysts were synthesized through the impregnation (Fe2O3/t-ZrO2) and the hydrothermal (Fe2O3/c-ZrO2) methods. The hydrothermal preparation improved the catalyst activity and propylene yield; propane conversion: 45.8 % vs. 31.2 %, propylene yield: 31.1 % vs. 23.2 %. This higher activity was attributed to the high specific surface area and thereby higher Fe dispersion that led to more formation of active Fe3 + sites. According to results, the t-ZrO2 surfaces were known to be more favorable for coke reaction compared to c-ZrO2. Adsorption energy of propane and propylene onto the most stable surfaces of ZrO2 was measured as a criterion for side reactions and coke deposits by DFT calculations. The results showed that propane and propylene are more adsorbed onto t-ZrO2 than onto c-ZrO2, which indicates more cracking of propane and propylene onto t-ZrO2. Finally, the experimental data were successfully fitted with the Deactivation Model with Residual Activity (DMRA). It was found that the Fe2O3/c-ZrO2 catalyst has a higher resistance against the deactivation by coke deposits compared to the Fe2O3/t-ZrO2 catalyst; activation energy of deactivation function: 83.64 vs. 70.09 kJ/mole.

Performance and Reaction Mechanism of a Novel Coking Binder Obtained from Low-Grade Coal.

Under the premise of ensuring the quality of coke, reducing coking/fat coal utilization is an urgent problem in the coking industry. This study prepared an extract with softening and bonding characteristics from noncaking low-grade coal by a degradative solvent extraction method. The substitution effect of coking/fat coal by the extract for coking was studied, and the effect mechanism was analyzed in detail. The results show that the caking index of the extract reached 100 and the vitrinite content reached 80%. When the extract substitution ratio is 10%, the coke quality is basically unchanged. Especially when the substitution ratio is 5%, all of the properties of coke are improved, the particle coke reactivity index (PRI) decreases by 4%, the particle coke strength after reaction (PSR) increases by 5%, and the coke microstrength index (MSI) increases by 12%. The influence mechanism model of extract addition on the strength and structure of coke was proposed. The lower softening point of the extract can broaden the plastic interval and improve the caking properties and fluidity of the blended coal, reduce the porosity of coke, and improve the order degree of the carbonaceous structure. Moreover, the highly reactive extract undergoes cross-linking reactions with coal during coking; thus, the cross-link bonds further improve the strength of the coke.

Modification mechanism of caking property of polystyrene waste using low-temperature pyrolysis and its use in coal-blending coking

Polystyrene (PS) waste was depolymerized using a low-temperature pyrolysis treatment (LTPT) to increase its caking index. The mechanism of caking index modification was revealed by using Fourier transform infrared spectroscopy, thermogravimetric (TG) analysis, pyrolysis-gas chromatography with mass spectrometric detection, and solid-state 13C nuclear magnetic resonance spectroscopy. The crucible coal-blending coking tests were carried out using an industrial coal mixture and the treated-PS with the highest caking index (PS300) or raw PS. Some properties of the resultant cokes were also analyzed. It was demonstrated that the caking index of PS dramatically increased by LTPT; however, exceeding 300 °C did not yield any benefit. The caking index increased due to the formation of the caking components, whose molecules are medium in size, caused by LTPT. Additionally, the coke reactivity index of the coke obtained from the mixture containing PS300 decreased by 5.1% relative to that of the coke made from the mixture with PS and the coke strength after reaction index of the former increased by 7.3% compared with that of the latter, suggesting that the ratio of depolymerized PS used for coal-blending coking could increase relative to that of PS.

Introducing K to Regulate the Electronic Structure of Perovskite-Type Oxide for the Efficient Reaction of Coke and Steam

Introducing K to Regulate the Electronic Structure of Perovskite-Type Oxide for the Efficient Reaction of Coke and Steam

A novel dual-way inference modeling method for coal coking: Predicting H2 and CH4 concentrations in coke oven gas and inferring optimal reaction conditions

Coal coking is an efficient and environmentally friendly technology for energy utilization that yields various industrial materials. However, the intricate nature of the coal coking reaction poses a challenge for chemical theories to fully capture and derive its reaction process. To address this challenge, a novel dual-way inference modeling method was proposed to simulate the coal coking process. This modeling method not only predicts H2 and CH4 concentrations, but also infers optimal operating conditions based on the desired H2 and CH4 concentrations. In this work, three representative machine learning methods (XGBoost, Random Forest, and ANN) are compared with dual-way inference modeling method, and the key factors affecting H2 and CH4 concentrations are thoroughly explored. The experimental results show that the primary factor influencing CH4 concentration is C, H2 concentration is predominantly affected by fixed carbon, while C plays the most important role for CH4 + H2. The dual-way inference modeling method achieved excellent performance in predicting H2 and CH4 concentrations and operating conditions. In the prediction task of H2, CH4, and H2 + CH4 concentrations, the coefficient of R2 achieved 0.9767, 0.9936, and 0.9851, respectively. The result for predicting coking temperature based on the desired H2 and CH4 concentrations indicate a positive correlation between coking temperature and both H2 and CH4 concentrations. In general, as coking temperature increases, the concentrations of both H2 and CH4 will increase.

Application of High‐ and Low‐Reactivity Cokes in Hydrogen‐Rich Blast Furnaces

The effects of two cokes with different reactivity on the lump ore's metallurgical properties and coke's solution loss are investigated under the high‐temperature load reduction. The work used an improved test device for softening‐melting and dropping characteristics of iron ores in both CO2 and CO2H2O atmospheres. The deterioration behavior of highly reactive cokes is expounded under hydrogen‐rich conditions. High‐reactivity cokes under hydrogen‐rich conditions are more favorable for enhancing the breathability of charge and the penetration of the coke layer. However, it increased the thickness of the softening zone. High‐reactivity cokes had obvious internal and external reaction gradients. The solution loss reaction mostly occurred on the surface, with selectivity. The longitudinal stacking height, layer number, and order degree in the carbon structure decreases after the reaction. The carbon‐structure difference weakens between the shell and core. The enhancement of coke's reactivity, however, results in the significant loss of coke powders on its surface. Unreduced FeO and refractory Fe2SiO4 are more likely to appear in the droplets, which is not conducive to the reduction of Fe and the generation of slag crust in the furnace. The difficulty in separating lump ores and cokes is aggravated, and more iron‐containing charge remain in the furnace.

A semi-industrial investigation about the use of iron-steel production by-products in reactive coke production

This study investigates the utilization of by-products from iron-steel integrated plants in the production of reactive coke. Various by-products were evaluated, and blast furnace flue dust with high iron (Fe) content was incorporated into the coal blend through the briquetting of highly volatile coal. The blend consisted of 10 % briquette coal comprise 1 % blast furnace flue dust. The addition of blast furnace flue dust with briquettes aimed to mitigate the reduction in coke strength caused by increased reactivity. Reactive coke was successfully produced through the evaluation of blast furnace flue dust, a by-product of iron and steel plants. The resulting reactive coke exhibited a reactivity value approximately 20 % higher than coke produced without the addition of blast furnace flue dust while maintaining its strength. By re-evaluating blast furnace flue dust as a by-product of the iron and steel process, the reactivity of coke increased without compromising its strength.

Study on using waste biomass as carbon reducing agent in industrial silicon smelting

In this study, seven types of biomass (coffee husk, macadamia husk, peanut husk, rice husk, rice straw, tobacco straw, and cotton straw) were carbonized at temperatures ranging from 300 °C to 1000 °C. This study explored the potential of these biomass samples as carbon reducing agents for industrial silicon and compared them with charcoal in terms of their proximate analysis, electrical resistivity, and apparent activation energy of the reaction. The results indicated that as the pyrolysis temperature increased, the oxygen content and aliphatic functional groups in biochar gradually decreased, as did the volatiles content, while the proportions of fixed carbon and ash increased. The resistivity decreased upon the removal of volatiles and an increase in the degree of graphitization. At 600 °C, the fixed carbon content of all biochar exceeded 65%, and the fixed carbon content of macadamia husk biochar reached 91.32%, which was higher than that of charcoal (81.44%), the maximum specific surface area was 335.706 m2/g, which was much higher than that of charcoal (5.705 m2/g). Thermogravimetric analysis and kinetic analysis indicated that coffee husk had better reactivity than charcoal. The gasification reactivity of coffee husk was R = 3.1949%/min °C, which was higher than that of charcoal (0.7930 %/min °C). The apparent activation energy during the coking reaction stage was 51.86 kJ/mol, which was lower than that of charcoal (98.9 kJ/mol). Overall, biochar had a suitable fixed carbon and electrical resistivity at 600 °C and can be used as a carbon reducing agent for industrial silicon, among them, macadamia husk with high fixed carbon and coffee husk with high reactivity have greater application potential.

The Perspective of Using Neural Networks and Machine Learning Algorithms for Modelling and Forecasting the Quality Parameters of Coking Coal—A Case Study

The quality of coking coal is vital in steelmaking, impacting final product quality and process efficiency. Conventional forecasting methods often rely on empirical models and expert judgment, which may lack accuracy and scalability. Previous research has explored various methods for forecasting coking coal quality parameters, yet these conventional methods frequently fall short in terms of accuracy and adaptability to different mining conditions. Existing forecasting techniques for coking coal quality are limited in their precision and scalability, necessitating the development of more accurate and efficient methods. This study aims to enhance the accuracy and efficiency of forecasting coking coal quality parameters by employing neural networks and artificial intelligence algorithms, specifically in the context of Knurow and Szczyglowice mines. The research involves gathering historical data on various coking coal quality parameters, including a proximate and ultimate analysis, to train and test neural network models using the Group Method of Data Handling (GMDH). Real-world data from Knurow and Szczyglowice mines’ coal production facilities form the basis of this case study. The integration of neural networks and artificial intelligence techniques significantly improves the accuracy of predicting key quality parameters such as ash content, sulfur content, volatile matter, and calorific value. This study also examines the impact of these quality indicators on operational costs and highlights the importance of final indicators like the Coke Reactivity Index (CRI) and Coke Strength after Reaction (CSR) in expanding industrial reserve concepts. Model performance is evaluated using metrics such as mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). The findings demonstrate the effectiveness of these advanced techniques in enhancing predictive modeling in the mining industry, optimizing production processes, and improving overall operational efficiency. Additionally, this research offers insights into the practical implementation of advanced analytics tools for predictive maintenance and decision-making support within the mining sector.

A comprehensive investigation of thermal coke formation during rapid non-catalytic pyrolysis of rubber seed oil

Due to coking deposits, catalyst deactivation severely affects the upgrading of triglyceride biomass. Understanding the chemical nature and development of coke species is essential for mitigating the degree of coking and for the regeneration process of the catalysts. In this study, SEM, XRD, FT-IR, Raman and XPS are employed to gain insights into the coking behavior of RSO during the rapid pyrolysis process. The results show that pyrolysis of RSO produces high quality syngas with 98.72 % H2 and CO. The unstable O-containing functional groups in the cokes continue to be lost, which indirectly promotes the condensation reaction of aromatic hydrocarbons. The hydroxyl group is the main functional group that affects the reactivity of cokes, and it is one of its more active groups. Although the crystal size of the cokes is growing, its degree of graphitization is decreasing with the temperature increases. The size of the aromatic ring system in the cokes gradually increases with increasing temperature. The cross-linking structure in the coke is constantly being destroyed. Resulting in some O-containing functional groups entering the cokes. Based on this work, the coke evolutionary route of rapid pyrolysis of RSO is proposed.

Characterization of Porosity Changes during Reaction of Coke and Coke Analogues with CO2 Using Micro‐Computer Tomography

Understanding the porosity of coke and how it changes during reaction is key for understanding coke reactivity and minimizing coke consumption in the blast furnace. The change in porosity of coke and coke analogue samples after reaction with CO2 is assessed using high‐resolution micro‐computer tomography (micro‐CT) scans. High‐resolution micro‐CT scans (2.35 μm voxel−1) are carried out before and after reaction with CO2 at 1100 °C. 3D micro‐CT measurements are validated against standard 2D measurements, and used to quantitatively assess the porosity evolution in the samples. The porosity and pore connectivity of the coke and analogue samples increase in reaction with CO2. Most of the increase in porosity occurs in the pores &lt;60 μm. Inert maceral‐derived component (IMDC) regions in coke A are more reactive than reactive maceral‐derived component (RMDC) regions. This is possibly related to the higher proportion of fine pores in unreacted IMDC compared to RMDC. Characterization of larger coke and coke analogue samples using micro‐CT at lower resolution (9.65–16.72 μm voxel−1) allows the rate‐controlling mechanism to be determined as likely to be mixed control, with both chemical reaction and pore diffusion‐controlling reaction rate. This finding demonstrates that high‐ and lower‐resolution micro‐CT datasets should be considered complementary.

The employment of domain adaptation strategy for improving the applicability of neural network-based coke quality prediction for smart cokemaking process

Precise coke quality prediction is essential for coke production process optimization to achieve the reduction in energy consumption and CO2 emissions, thus moving toward carbon neutrality in the coking industry. However, the complexity of coal molecular structures and the chemical reactions in the cokemaking process pose significant challenges to the applicability of existing coke quality prediction models. Based on the recently widely employed Artificial Neural Network (ANN) method, this study is the first to introduce domain adaptation strategies for improving the applicability of ANN in predicting coke quality including Coke Strength after Reaction (CSR) and Coke Reactivity Index (CRI). 649 Chinese coal samples with properties including Mad, Ad, Vdaf, St,d, G, X and Y along with coke quality were collected. They were initially categorized into source and target domains characterized by different distributions. Subsequently, two scenarios were independently evaluated based on coal samples in the target domain, which either lacked any information or contained partial information of actual coke quality. The results suggested that the proposed approach can significantly enhance the predictive performance for coal samples across various distributions. Moreover, a comprehensive investigation was also conducted to determine key factors influencing the effectiveness of coke quality prediction with this approach.

Co-pyrolysis of coal with biomass residues and coke breeze for superior quality coke via hot pressing technology

Superior quality coke is essential in the metallurgical industry. However, the coking coal source is limited. The biomass or breeze could be blended into the coking coal to produce high-standard cokes by hot pressing technology. High-standard cokes should have low reactivity (chemical reaction intensity of coke and CO2 at high temperature) and high strength. The influences of biomass category and blending ratio on pore structure, reactivity and crushing strength of cokes were explored. Pore evolution during the coking process was studied by experiments and simulations to reveal the mechanism of producing high-quality cokes with biomass and breeze blended into coking coal via hot pressing technology. Results show that hot pressing technology can make high-quality cokes when the blending ratio of biomass or breeze is appropriate. When 5% bamboo, wood or straw is blended to produce hot-pressed coke, the coke reactivity decreases by 6.58%, 7.66% and 6.41% and the crushing strength increases to 1.15, 1.19 and 1.12 times when compared with coke in case of no blending and no pressing, respectively. The proper blending ratio of the three biomasses should be less than 15% because the coke quality deteriorates greatly when the blending ratio increases to 15%. If the breeze is mixed with coal to produce high-standard cokes under hot pressing conditions, the blending ratio should not exceed 30%. Hot pressing technology can reduce the coke pores when the biomass or breeze is blended into the coal with a low ratio. When 5% bamboo or 10% breeze is blended, the porosity of hot-pressed coke in the center region decreases by 5.69% and 4.02% than coke with no blending and no pressing, respectively. The low porosity contributes to the reduction of coke reactivity. The fewer pores, thicker pore walls and higher carbon matrix strength impede the growth of cracks, resulting in a higher coke crushing strength. Therefore, hot pressing technology can effectively enhance coke quality if the biomass or breeze is mixed with the coking coal with an appropriate blending ratio.

Optimizing the performance of iron coke by coal blending: Insights from the metallurgical properties and structural characteristics

Iron coke has gained widespread attention for its advantages in carbon reduction and efficient resource utilization. However, the structure and strength of iron coke have constrained its further advancement. This study systematically explores the impact of different coal blending structures on the three-dimensional structure and tensile strength of iron coke through three-dimensional reconstruction technology. It highlights the improvement mechanism of 1/3 coking coal addition on the microstructure and metallurgical properties of iron coke. The study shows that the tensile strength of iron coke increases with the addition of coking coal, while lean coal exhibits the opposite trend. The addition of an appropriate proportion of 1/3 coking coal (20–22 %) enhances the tensile strength and mechanical strength of iron coke, reaching the level of second-grade coke. However, an addition exceeding 22 % leads to increased porosity and reduced strength due to the high volatile content. Finite element analysis demonstrates that stress is primarily localized in regions with thinner pore walls and at the iron-carbon interface, and the aggregation effect of iron particles further causes stress concentration. Iron coke reactivity initially diminishes and subsequently rises with the addition of 1/3 coking coal, and it shows a linear negative correlation with the strength after reaction. Structural evolution reveals that the addition of 1/3 coking coal reduces the orderliness of carbon microcrystals, thereby increasing the active sites for gasification reactions. Meanwhile, the initial reduction followed by an increase in the specific surface area of iron coke can impact adsorption sites, thereby influencing reaction activity. These findings not only provide a new understanding of the structure-performance relationship of iron coke but also offer practical guidance for iron coke production.

The Influence of Organic and Inorganic Additives on the Specific Electrical Resistance of Coke

This study aimed to evaluate the effect of both inorganic (boron carbide nanopowders and silicon carbide (carborundum) and organic lean (petroleum coke) additives on the quality of coke produced in a laboratory furnace, as well as on its electrical properties. Analyzing the results of the quality assessment of the obtained coke, it can be argued that the addition of a fixed amount (0.25-0.5 wt.%) of non-caking nanoadditives allows to regulate the process in the plastic state in order to increase the coke strength. This modification affects the coke quality and has a significant dependence on the grade composition of the coal charge. The use of nanoadditives is especially important for coal charges with poor coking properties. Adding 5% of petroleum coke to the coal charge leads to an increase in the gross coke yield by 1.2-1.3%; a decrease in coke ash content by 0.2-0.3%; an increase in the total sulfur content in coke by 0.15-0.23%; deterioration in both mechanical (P25 − by 0. 1-0.6%; I10 − by 0.1-0.2%) and coke strength after the reaction (CSR - by 0.6-1.0%), coke reactivity (CRI - by 0.2-0.3%), as well as structural strength (SS by 0.3-0.4%), abrasive hardness (AH by 0.7-1.0 mg) and specific electrical resistance (ρ by 0.002-0.007 Om×cm). The obtained data may indicate an increase in the order degree of the coke structure and the appearance of a larger number of nanostructures. In addition, it should be noted that a sharper deterioration in blast furnace coke quality is observed when using a coal charge characterized by a lower coal content of the Concentrating Factory Svyato-Varvarynska LLC.

Selective oxidation of FeNiCrAl-based alloys under low oxygen pressure and their coking resistance

Selective oxidation of Fe–25Ni–20Cr–xAl-based alloys under low oxygen pressure was studied to improve oxidation and coking resistance. The morphology and oxidation kinetics of the oxide layer were analyzed using multi-scale techniques such as scanning electron microscopy, electron probe micro-analysis, and X-ray photoelectron spectroscopy. The results indicated that when the Al content exceeded 2.5 wt%, a duplex-structure oxide film, mainly consisting of M2O3 (M: Al and Cr) outer and Al2O3 inner layer, was formed on the surface of the alloy after pre-oxidation. When the Al content exceeded 8 wt%, a complete and dense Al2O3 external oxide film was formed through a two-step oxidation process. The decrease in oxygen partial pressure and surface enrichment of aluminum during the pre-oxidation process accelerated the formation of an external Al2O3 oxide film. Therefore, it isolated the catalytic coking reaction of Fe and Ni transition metals in the alloy with the pyrolysis atmosphere, significantly improving the anti-coking performance of the alloy.

Study on the High-Temperature Interaction between Coke and Iron Ores with Different Layer Thicknesses.

Coke plays a key role as the skeleton of the charge column in BF. The gas path formed by the coke layer in the BF has a decisive influence on gas permeability. At high temperatures, the interface between coke and ore undergoes a melting reaction of coke and a reduction reaction of ore. The better the reducibility of the ore, the more conducive it is to the coupling reaction of ore and coke. The melting loss reaction of coke becomes more intense, and the corresponding strength of coke will decrease, which will affect the permeability of the blast furnace and is not conducive to the smooth operation of the blast furnace. Especially with a deterioration in iron ore quality, BF operation faces severe challenges, which makes it necessary to find an effective way to strengthen BF operation. In this study, a melting-dropping furnace was used to develop and clarify the high-temperature interaction between coke and iron ores with different layer thicknesses. The influencing factors were studied by establishing a gas permeability mathematical model and observing the metallographic microscope images of samples after the coke solution loss reaction. The relationships between coke layer thickness, distribution of gas flow, and pressure drop were obtained. The results showed that, under certain conditions, the gas permeability property of a furnace burden has been improved after the coke layer thickness increased. Upon observing the size of coke particles at the interface reaction site, the degree of melting loss reaction can be determined. A smaller particle size indicates more melting loss reaction. A dripping eigenvalue for molten metal was introduced to evaluate the dynamic changes in the comprehensive dripping properties of molten metal of furnace burden, which showed that the dripping eigenvalue for the molten metal could deteriorate because of the unruly thickness and the coke layer thickness should be limited through considering the operational indicators of the blast furnace.

A novel image expression-driven modeling strategy for coke quality prediction in the smart cokemaking process

In pursuit of carbon neutrality and advancing energy-efficient practices within the steel and coking industries, the traditional cokemaking process is progressively evolving towards intelligence, with coke quality prediction emerging as a pivotal technology at its core. Nevertheless, the intricacy of the coke production process presents a formidable challenge in accurately forecasting it. This study is the first to propose a novel image expression-driven modeling approach that transforms the numerical coal properties into image expressions and uniquely integrates the utilization of the Convolutional Neural Network (CNN) for predicting coke quality including Coke Strength after Reaction (CSR) and Coke Reactivity Index (CRI). Utilizing the collected 729 Chinese coal properties and corresponding coke quality indexes, the dimensionality reduction technique was employed to transform numerical coal properties into image expressions. A convolutional neural network combined with the random forest model was subsequently developed for learning and prediction, with its performance evaluated on Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R2 metrics. The results suggested that the proposed groundbreaking model outperformed existing numerical properties-based coke quality prediction models and typical regression models, achieving MAE of 1.57, RMSE of 2.22, and 0.86 for R2 metric, along with MAE of 1.82 and RMSE of 2.42 as well as 0.91 for R2 metric in CRI and CSR prediction, respectively. Furthermore, a comprehensive analysis was also undertaken to identify the pivotal factors influencing the efficacy of coke quality prediction based on the proposed approach.