In this paper, we extend a recent asymptotic axisymmetric model for isothermal gas–solid flow in a countercurrent moving-bed reactor to a non-isothermal model for the heat transfer in an ironmaking blast furnace. The appended heat transfer model accounts for conduction, convection, thermal non-equilibrium between the gas and solid phases and the dominant endothermic chemical reaction in the bulk of the furnace, the Boudouard reaction. Asymptotic analysis is used to determine the leading-order heat balances and to interpret numerically obtained solutions for the phase temperatures. Although the model is considerably simpler than the many numerical models that already exist for blast-furnace operation, its future extension would form the basis of a computationally efficient approach for modelling the transient state of a blast furnace.

Macroscopic interactions of liquid iron and solid oxides, such as alumina, calcia, magnesia, silica, and zirconia, manifest the behavior and efficiency of high-temperature metallurgical processes. The oxides serve dual roles, both as components of refractory materials in submerged entry nozzles and also as significant constituents of nonmetallic inclusions in the melt. It is therefore crucial to understand the physicochemical interplay between the liquid and the oxides in order to address the nozzle clogging challenges and thereby optimize cast iron and steel production. This paper presents a methodology for describing these interactions by combining the materials' dielectric responses, computed within the density functional theory, with the Casimir-Lifshitz dispersion forces to generate Hamaker constants. The approach provides a comprehensive understanding of the wettability of iron against these refractory oxides, revealing the complex relation between the molecular and macroscopic properties. Our theoretically determined crystalline structures are confirmed by room-temperature X-ray diffraction, and the contact angles of liquid iron on the oxides are validated with a sessile drop system at a temperature of 1823 K. For comparison, we also present the wettability of the oxides by a liquid tin-bismuth alloy. The findings are essential in advancing the fundamental understanding of interfacial interactions in metallurgical science and pivotal in driving the development of more efficient and reliable steelmaking processes.

In this paper, we investigate the Casimir-Lifshitz free energy mechanism that governs both ice growth and melting near metal surfaces, with a particular focus on the role of oxidation. Our study reveals that metals such as gold, iron, and aluminum induce incomplete premelting, resulting in micron-sized liquid water layers when in contact with ice. These layers could have significant implications for the defrosting of metallic surfaces. When exposed to water vapor at the triple point, aluminum and other metals can induce the formation of notably thick layers of either liquid water or ice, which can theoretically become infinitely thick if other interactions are disregarded. However, when aluminum undergoes oxidation to form alumina, its behavior changes dramatically. Alumina surfaces cause complete melting when in direct contact with bulk ice and result in only micron-sized layers of water or ice in vapor conditions. In contrast, magnetite, the oxidized form of iron, retains metalliclike behavior due to its high dielectric constant, similar to other metals, and continues to support thick layers of water or ice. This distinction highlights the significant influence of oxidation on the dynamics of ice growth and melting near different metal surfaces.

In the blast furnace and basic oxygen furnace route, pig iron and steel scrap are used as resources for steel production. The scrap content can consist of many different types of scrap varying in origin and composition. This makes it difficult to compile the scrap mix and predict the future chemical analysis in the converter. When compiling the scrap mix, steel manufacturers often rely on experience and trials. In this paper, we present a machine learning approach based on XGBoost to predict the chemical element content in the converter. Data from around 115000 heats were analyzed and a model was developed to better predict the content of the tramp elements copper, chromium, molybdenum, phosphorus, nickel, tin and sulphur at the end of the basic oxygen furnace process. The study shows that it is possible to predict the chemical element content for tramp elements in the converter based solely on data available in advance and routinely collected without the necessity of additional sensors or analysis of input material. Given the nature of scrap classifications for (external) scrap types, this is non-trivial. Furthermore, an online model was implemented, accessible via a defined synchronous interface, which allows to optimize the use of different scrap types by predicting the chemical content at the end of the basic oxygen furnace process and simulating with new combinations of input material. Not all types of steel scrap are always available. With the model developed, new scrap input constellations can now be created to ensure that the quality of the melt is maintained. However, for very accurate predictions, the data from the upstream processes must be of high quality and quantity. Efficient scrap management, monitoring of the scrap input and confusion checks.

The operational efficiency of continuous casting in steel production is often hindered by the clogging of submerged entry nozzles (SEN), caused due to the agglomeration of nonmetallic inclusions (NMIs). SEN clogging is challenging to monitor and requires probabilistic models for accurate real-time prediction. In this context, data-driven models emerged as a promising tool to be used in the existing industrial settings. Despite frequent occurrence of SEN clogging, collecting large datasets under varied operational conditions remains challenging. The scarcity of data hampers the ability to develop and train traditional data-driven models effectively. To overcome these challenges, physics-informed data-driven models are proposed in this work. The integration of outputs generated from theoretical calculations is sufficient to compensate for the lack of available datasets. To further enhance accuracy, an advanced methodology involving use of ab initio repository is developed. This repository contains material-specific data including high-temperature nonretarded Hamaker constants of NMIs in specific particle size range of 1–10 μm. A novel parameter, “Clogging Factor” is proposed to monitor and integrated into the modeling architecture to track the reduction in the available volume inside SEN due to the accumulation of NMIs. The proposed model has yet to be validated online but has shown potential in reducing SEN clogging.

Continuous casting of aluminum (Al) deoxidized steels demands careful inspection due to the occurrence of submerged entry nozzle (SEN) clogging, leading to unexpected production stops. Recognizing the castability of aspecific “cast” by monitoring the condition of the SEN is essential for uninterrupted casting. With this information prior to casting, operators can take preventive action against possible clogging occurrences, thus reducing unplanned downtimes. In response to the severe implications of SEN clogging, this work introduces a novel way to forecast castability of Al-killed steels. A hybrid model is proposed that integrates the adaptive neuro-fuzzy inference system (ANFIS) and long short-term memory (LSTM) networks. The output of the model helps to anticipate the event of clogging by analyzing both the past condition of the SEN and changes in the steel chemistry during the transport of the steel ladle from refining to the casting process. A comprehensive analysis of 150 casts helped to build the ANFIS algorithm for estimating the castability index (CI) parameter from steel chemistry. LSTM algorithm is used as asubsequent step to forecast castability in the next 20–25 min. Discrepancies between the predictive response and the actual conditions are reported. Although the real-time implementation of the proposed model is the ultimate goal, the focus of this work was to present the methodology and demonstrate its potential.

Melt infiltration into porous media is an intriguing phenomenon that holds immense significance across various sciences and technologies. In this work, the problem of metallic melt infiltration in particulate beds is investigated for understanding and prediction of severe accident progression associated with a molten pool penetrating through an underlying debris bed which may form in the reactor core or in the lower head of a light water reactor. The present study aims to quantify the effect of particle surface's wettability on melt infiltration kinetics. For this purpose, two categories of experiment are conceived and carried out to measure the wettability of different material surfaces by melt and to characterize melt infiltration kinetics in one-dimensional particulate beds, respectively. The melt material is tin–bismuth eutectic alloy with a melting point of 139 °C. Copper (Cu), stainless steel (SS), Tin (Sn) and tin-coated stainless steel (Sn-coated SS) are chosen as materials of substrates and particles in wettability measurement and melt infiltration study. The particulate beds, packed with 1.5 mm spheres, are preheated to 200 °C before the melt infiltration begins. The experimental data of wettability measurement shows that the contact angles of liquid Sn-Bi eutectic on the above-mentioned material surfaces range from 79° to 135°. The results of melt infiltration tests confirm the significant effect of wettability on melt penetration kinetics. The capillary force plays a significant role in the initial infiltration of particulate beds. Specifically, a wettable particulate bed enhances the initial melt infiltration, whereas non-wettable beds hinder it.

There has been remarkable progress in the field of Deep Learning and Computer Vision, but there is a lack of freely available labeled data, especially when it comes to data for specific industrial applications. However, large volumes of structured, semi-structured and unstructured data are generated in industrial environments, from which meaningful representations can be learned. The effort required for manual labeling is extremely high and can often only be carried out by domain experts. Self-supervised methods have proven their effectiveness in recent years in a wide variety of areas such as natural language processing or computer vision. In contrast to supervised methods, self-supervised techniques are rarely used in real industrial applications. In this paper, we present a self-supervised contrastive learning approach that outperforms existing supervised approaches on the used scrap dataset. We use different types of augmentations to extract the fine-grained structures that are typical for this type of images of intrinsically unordered items. This extracts a wider range of features and encodes more aspects of the input image. This approach makes it possible to learn characteristics from images that are common for applications in the industry, such as quality control. In addition, we show that this self-supervised learning approach can be successfully applied to scene-like images for classification.

The substitution of fossil coal with biocarbon in the metallurgical processes can help to decrease fossil CO₂ emissions. Biocarbon’s characteristics, such as high volatile matter contents and high reactivities with CO₂, are beneficial for increasing the reduction degrees and reduction rates of iron oxides in carbon composite agglomerates (CCA). This study compared the reduction of hematite by of two types of carbonaceous materials (CM): hydrochar (high-volatile biocarbon) and anthracite (a low-volatile coal) in the form of CCA. CM, hematite, and binder (starch) were mixed together to obtain mixtures with C/O molar ratios equal to 0.4–1.2. The mixtures were reduced non-isothermally in nitrogen atmosphere up to 1003 K or 1373 K. Up to 1003 K, the volatiles released from CMs and starch reduced hematite by 18–35%. Between 1003 K and 1373 K, both hydrochars (produced from lemon peels and rice husks) reacted with iron oxides more rapidly than anthracite below 1360 K, when the samples had C/O ratios in the range of 1.0–1.2. In this temperature range, rice husk hydrochar promoted a slower reaction with iron oxides than lemon peel hydrochar, which was possibly influenced by its higher ash content which decreased the rate of Boudouard reaction. Samples with C/O ≥ 1.0 achieved complete reduction at 1373 K, regardless of the type of CM used, whereas samples with C/O equal to 0.4–0.5 achieved 63–86% reduction. It can be concluded from this study that hydrochar can fully substitute anthracite for direct reduction of iron oxide to decrease fossil CO2 emissions during ironmaking processes.

For today’s high-quality steel production, good control of slag composition isessential in secondary steelmaking. However, the conventional slag analysispractice, involving sampling, sample preparation, and analysis, is very timeconsuming.This work is the first step toward an investigation of infrared (IR)-basedsystems and can be used for online slag composition monitoring using theprinciple that different slag compositions have different emissivities in the IRwavelength range. Therefore, this work experimentally determines emissivityvalues of slags as a function of composition at steelmaking temperature, sinceavailable data for slags are very limited in the literature. The emissivities of threedifferent slag compositions belonging to the Al2O3–CaO–SiO2–MgO system areinvestigated at 1773 K. The investigated emissivities are in the range of 0.75–0.87,with the best repeatability seen in the slag which is fully liquid at 1773 K. Variationsin emissivities are observed within the other slags due to the presence of solidphases. Although the data clearly indicate a difference of emissivities as a functionof slag composition, further experiments must be performed to evaluate theemissivities of other characteristic slags at different temperatures in order tofurther assess the applicability of IR-based systems for slag composition control.