Chord recognition serves as a critical task in music information retrieval due to the abstract and descriptive nature of chords in music analysis. While audio chord recognition systems have achieved significant accuracy for small vocabularies (e.g., major/minor chords), large-vocabulary chord recognition remains a challenging problem. This complexity also arises from the inherent long-tail distribution of chords, where rare chord types are underrepresented in most datasets, leading to insufficient training samples. Effective chord recognition requires leveraging contextual information from audio sequences, yet existing models, such as combinations of convolutional neural networks, bidirectional long short-term memory networks, and bidirectional transformers, face limitations in capturing long-term dependencies and exhibit suboptimal performance on large-vocabulary chord recognition tasks. This work proposes ChordFormer, a novel conformer-based architecture designed to tackle structural chord recognition (e.g., triads, bass, sevenths) for large vocabularies. ChordFormer leverages conformer blocks that integrate convolutional neural networks with transformers, thus enabling the model to capture both local patterns and global dependencies effectively. Experimental results show that ChordFormer achieves a 2.3% improvement in frame-wise accuracy and a 6% increase in class-wise accuracy on large-vocabulary chord datasets. Moreover, our evaluation demonstrates that ChordFormer handles class imbalance more effectively than existing models, achieving balanced recognition across a wide range of chord types. We further assess the ChordFormer effectiveness through detailed ablation studies and additional metrics, including ARWCSR and chord inversion matrices. This approach bridges the gap between theoretical music knowledge and practical applications, advancing the field of large-vocabulary chord recognition.

Managing environmental noise is a critical challenge for industries near residential areas. This study presents a data-driven framework for real-time monitoring and control of acoustic emissions in a large-scale steelworks. A multi-stage approach for proactive acoustic governance is proposed, which integrates process and meteorological data. XGBoost classification models ensure regulatory compliance by effectively detecting overcoming of noise limit. Continuous noise estimation is addressed via Deep Learning: while a Standard Feedforward Network provides high predictive accuracy, an innovative Architecture-Aware Model demonstrates that embedding domain knowledge achieves comparable performance with a significant reduction in parameters, increasing robustness. Finally, Self-Organizing Maps offer an exploratory interface to identify quieter operational regimes through topology-preserving mapping. This integrated AI framework provides a powerful Decision Support System for industrial noise mitigation and operational optimization.

At near-atmospheric pressure, Devanathan–Stachurski (DS) permeation measurements operate near reversible-trap saturation, preventing separate identification of trapping parameters. We show that sub-atmospheric H2 partial pressures (0.01 to 1.00 bar) allow the reversible-trap density Nr and a model-derived effective binding energy Eb to be jointly identified from multi-pressure data. Two advances enable this regime in a hybrid gas/electrochemical DS cell: (i) a flow-mixed Ar/H2 inlet controlling partial pressure by mass-flow ratio, and (ii) an in-cell palladium electroplating procedure under real-time monitoring that suppresses blistering on thin ARMCO iron foils. Measured steady-state fluxes scale as the square root of partial pressure, validating Sieverts’ law. The 6-pressure dataset is described by the Johnson model (R2=0.919), yielding Nr=1.17⋅1017 cm−3 and Eb=59.3 kJmol−1, consistent with literature. Individual McNabb–Foster fits show larger parameter scatter, and a global McNabb–Foster fit sharing one trap-parameter set across all pressures provides an independent cross-validation of the Johnson result.

Electric steelworks are a paradigmatic representation of the concept of circular economy, as it recycles steel components at the end-of-life products. Moreover, its importance is foreseen to grow according to the increasing demand of decarbonizing steel production to meet the ambitious goals of the European Green Deal. The electric arc furnace-based route is still characterized by a limited diversification of energy supply sources thus, managing the three factors of smart energy management, energy prices, and production planning can be jointly considered as a crucial leverage for reducing production costs while ensuring satisfaction of energy demand coming from the different processes and developing digital approaches and tools to implement the fast adaptation to power grid behaviour. The article describes a novel energy management system based on innovative components and a flexible infrastructure, which uses machine learning and an optimization approach to minimize electricity consumption and level trends by matching intelligent production planning and power grid offer and related energy costs. The developed solution and the set of neural networks-based models estimating electricity consumption in electric arc furnace and ladle furnace based on production information are described. The models were trained and validated using production and process data from a real steelworks.

Although electric arc furnace (EAF)-based steelworks produce steel from recycled ferrous scrap and inherently implement the concept of circularity, they are challenged to reduce their overall environmental impact, reduce CO2 emissions, and maximize energy and resource efficiency. The paper exemplary shows how advanced digital technologies, including artificial intelligence-based techniques, can support decarbonization and sustainability improvement of electric steelmaking. The paper presents computationally efficient machine learning models estimating sterile content in different types of scrap reaching the scrap yard as well as steel chemical composition and temperature at the exit of the Ladle furnace. The models are designed to be included in an innovative software platform based on federated learning (FL) helping industrial staff in decision-making by estimating energy consumption and other parameters affecting the environmental impact according to the material mix fed to the EAF. The paper describes the rationale behind models' design, the approach for selecting their hyperparameters, and the results achieved on data gathered from two different steelworks, the first one exploited as reference for models' first setup, the second one used to assess models' usability in the considered FL context. The performances are satisfactory in both cases, and key issues for implementation and further improvement are discussed.

The European steel sector is committed to improve sustainability of the whole steel production chain, from decarbonisation of major upstream processes up to all downstream operations, including rolling. In particular, in the cold rolling process, oil-in-water emulsions are usually applied to lubricate the cold rolling process of low-Carbon steel. Such emulsions present some drawbacks mainly related to emulsion bath maintenance, subsequent production stages and waste disposal. Past research works showed that in some application areas, Oil Free Lubricants (OFL) show lubricant properties that are comparable to conventional lubricants, while providing significant environmental benefits. These lubricants are formulated as aqueous dispersions of Polyalkylene Glycols (PAG), a water-soluble synthetic polymer base, combined with various additives for lubrication enhancement, corrosion protection, and oil rejection. The project entitled "Transfer of aqueous oil free lubricants into steel cold rolling practice" (Ref. RollOilFree II - G.A. No 101112433) aims at developing an Oil-Free Lubricant for the cold rolling process of low-Carbon steel for applications in the automotive and packaging sectors by assessing its performance in industrial conditions. To this aim, the project combines tests at laboratory scale and simulations with trials in an industrial pilot cold rolling mill and, finally, field trials at industrial scale. Results demonstrate that OFL01 and OFL02 represent the most promising formulations as substitutes for commercial lubricants.The paper overviews the work undertaken in the first 18 months of the project, including laboratory investigations and part of the pilot trials.

Bio-CO2 is part of the natural carbon cycle and represents a sustainable carbon source for the production of Renewable Fuels of Non-Biological Origin (RFNBOs), such as synthetic methanol. This study addresses the critical knowledge gap in aligning diverse biogenic CO2 sources with e-methanol requirements in the EU by providing harmonized mapping, based on datasets, literature sources, and reported industrial statistics at the sectoral and country level. Bio-CO2 streams from biogas and biogas upgrading, biomass combustion, pulp and paper, bioethanol production, and the food and beverage sector are evaluated for total emissions, CO2 concentrations and purity, the geographical distribution, seasonality, and impurity profiles. Results show that approximately 350 Mtpa of bio-CO2 are emitted across the EU, with highly heterogeneous characteristics. Biogas upgrading and fermentation-based processes generate highly pure CO2 streams (>98–99%), yet their small and dispersed nature complicates logistics. In contrast, biomass-combustion and pulp and paper sectors provide large volumes (around 214.6–298.2 Mtpa and 73.9 Mtpa CO2, respectively), but in diluted streams (typically 3–15% and 10–20%). Replacing just 10% of the EU maritime fuel demand with e-methanol would require 53.6 Mtpa of bio-CO2 and 58 GW of electrolyzer capacity, a stark contrast to the current operational 385 MW. The findings highlight the need for infrastructure planning and aggregation hubs to enable the large-scale deployment of RFNBO methanol in the maritime sector.

Electric steelmaking is pivotal for the transition toward carbon-lean processes. Replacing fossil carbon and fuels with alternative nonfossil materials can contribute to enhancing the sustainability of this route. The investigations reported in this article explore the use of alternative carbon sources for slag foaming in the Electric Arc Furnace and the utilization of hydrogen in related burners. The effects of using alternative carbon sources are investigated via industrial trials and complementary simulations employing a flowsheet model of the entire electric route. The investigations demonstrate that, although alternative carbon materials can generally lead to fossil CO2 reduction of up to 15% without negatively affecting the product or most process aspects, high ratios of certain materials, such as 30% tires, can compromise operational safety and result in poor slag foaming. Concerning hydrogen use in burners, simulations show that CO2 reduction of up to 48% can be achieved in off-gases before post-combustion, accompanied by water vapor increases of up to 31%. Simulations also estimate an increase in hydrogen content in tapped metal of up to twice the reference value; however, this increase can be mitigated by standard vacuum degassing procedures.

Industrial Symbiosis refers to a collaborative approach, including synergies among companies for the transaction of resources, such as materials, energy, water, and by-products, thus resulting in mutual benefits and promoting a Circular Economy approach. Over the past decades, the steel sector was committed to reduce waste production as well as reuse waste and by-products to exploit them as a resource. Significant results in Industrial Symbiosis implementation have been achieved, creating new synergies and networks with other industrial sectors. Nonetheless, a comprehensive analysis of technical and non-technical barriers, that hinder the successful implementation of Industrial Symbiosis within the steel sector, can help implement an integrated and synergic approach encompassing and merging results and experience already achieved. This review paper presents a comprehensive overview of recent studies on the research trends on Industrial Symbiosis, considering drivers and barriers to its implementation, and maps the recent achievements related to the steel sector, by analysing the impact of some significant case studies. For instance, CO 2 valorisation in flue gases and steel slags to produce silicates and carbonates via mineral carbonation and CO 2 capture, re-use and sequestration by industrial symbiosis activities involving the steel and ammonia/urea industries are presented. The literature review based on selected publications allowed tracing the evolution of Industrial Symbiosis over the last few years. The assessment of main lines for research in Industrial Symbiosis allows identifying the challenges for future research. The analysis of implementation of new technologies can help to create new symbiotic networks and further developments and scenarios for the steel industry in a future characterized by material scarcity, decarbonization, and more stringent environmental legislation.

Stahl ist ein langlebiges Material und in allen Phasen des Produktlebenszyklus zu 100

Ensuring safety and operational efficiency in Electric Arc Furnace (EAF) steel manufacturing is critical due to the extreme hazards such as intense heat, toxic emissions, and heavy machinery present in these environments. We propose EAFvision, a real-time automated pipeline for safety surveillance EAFs, leveraging advanced deep learning architectures. EAFvision enables real-time detection of critical safety-related situations, including personnel, electrode clamps, and smoke emissions, to enhance situational awareness and operational safety in industrial environments. We collected and carefully annotated a comprehensive image dataset from an active EAF facility to benchmark a variety of models, including YOLO versions 8 through 11, RT-DETR, and established two-stage detectors like Faster R-CNN and Mask R-CNN. Our results demonstrate that lightweight, single-stage detectors deliver superior accuracy and faster inference times compared to more complex models, enabling efficient real-time testing on edge devices for immediate hazard detection and automated response. This approach highlights the transformative potential of AIpowered real-time monitoring systems to enhance workplace safety and optimize steel production processes.

The current, continuous increase in attention toward preservation of the environment and natural resources is forcing resource-intensive industries like steelworks to investigate new solutions to improve resource efficiency and promote the growth of a circular economy. In this context, electric steelworks, which inherently implement circularity principles, are spending efforts to enhance valorization of their main by-product, namely slags. A reliable characterization of the slag’s composition is crucial for the identification of the best valorization pathway, but, currently, slag monitoring is often discontinuous. Furthermore, in the current period of transformation of steel production, preliminary knowledge of the effect of modifications of operating practices on slags composition is crucial to assessing the viability of these modifications. In this paper, a stationary flowsheet model of the electric steelmaking route is presented; this model enables joint monitoring of key variables related to process, steel and slags. For the estimation of the content of most compounds in slags, the average relative percentage error is below 20% for most of the considered steel families. Thus, the tool can be considered suitable for scenario analyses supporting slag valorization. Higher performance is achievable by exploiting more reliable data for model tuning. These data can be obtained via novel devices that gather more numerous and representative data on the amount and composition of slags.

Direct reduction is considered a suitable alternative process for the transition of steelmaking route towards C-lean processes. If direct reduction is performed with high percentage of hydrogen use in reducing gas, carbon dioxide emissions are expected to significantly decrease. However, investigations of the effects of integration of hydrogen-enriched direct reduction processes in existing steelmaking routes are required to provide steel companies with valuable guidelines for selecting the most economically, technologically and environmentally sustainable transitions steps. Therefore, within the EU-funded project entitled "Maximise H2 Enrichment in Direct Reduction Shaft Furnaces" a first version of stationary flowsheet models of both Energiron-ZR and Midrex Direct Reduction processes were developed in Aspen Plus V14 (R). The models include shaft furnace and all auxiliary units and fit well with literature reference data. They will be used in a process chain multipurpose simulation toolkit to simulate the transition from standard integrated steelmaking route to a hydrogen-enriched direct reduction-based steelmaking route considering both production and gas and energy management aspects. In addition, these models will be the basis for dynamic models to investigate flexible operation of new integrated steelworks with hydrogen-enriched direct reduction.

Improving the sustainability of the steelmaking sector is a challenging task because steelmakers are expected to meet environmental targets and strict quality requirements that depend on the final application of the product. Recycling steel in electric arc furnaces (EAFs) is a well-established circular practice that helps reducing the environmental impact of steelmaking. However, an optimal combination of different scrap types and additions, along with minimum electricity and gas consumption during the melting phase, is necessary to ensure high quality and environmental performance of final products. The process input mix can be improved by exploiting optimization tools and Life Cycle Assessment (LCA) to minimize a multi-objective function including environmental impacts, constrained by technical requirements of steel and process operating conditions. The paper presents a methodology to transform “traditional” LCA into an “optimized LCA approach”, focusing on how Life Cycle Inventories and Life Cycle Impact Assessment can be associated to optimization variables or inputs, depending on steelmakers’ ability to affect EAF-based steelmaking operational parameters. The discussion highlights opportunities and limitations of integrating LCA and optimization methodologies within the framework of a real-world case study carried out in the European project ALCHIMIA.

The paper addresses the Coil-Order Allocation problem in steel industry via Genetic Algorithms through two approaches: a basic solution with a standard objective function and an advanced method incorporating a Fuzzy Inference System to mimic human decision-making. Both solutions were tested on real-world data from a tinplate production plant, achieving significant improvements in orders fulfillment and material utilization compared to manual allocation. The basic genetic approach outperforms the baseline in efficiency, while the fuzzy-genetic method demonstrate flexibility for complex, customizable optimization. The results show the potential of combining heuristic techniques and fuzzy logic to enhance industrial operations. (c) Copyright 2025 The Authors.

Oil-in-water emulsions (O/W emulsions) are generally used to lubricate the cold rolling process of low-carbon steel. In addition to the obvious advantages of efficient lubrication and cooling of the process, there are also some disadvantages, mainly related to emulsion bath maintenance, subsequent production steps and waste disposal. In some application areas, Oil-Free Lubricants (OFL’s) have been shown to be at least equally effective in decreasing friction and wear as conventional oil-based lubricants, while resulting in benefits related to waste disposal. In 2023, a project named “Transfer of aqueous oil free lubricants into steel cold rolling practice” (acronym ‘RollOilFreeII’) began, with it receiving funding from the Research Fund for Coal and Steel (RFCS). This project aims at an industrial application of Oil-Free Lubricants in the steel cold rolling process. The project builds on the work of the ‘RollOilFree’ project (also carried out in the RFCS-framework). This article briefly recapitulates the findings in the RollOilFree project and describes the objectives, benefits, activities and first results of the RollOilFreeII project. Notably, a pilot mill trial at high speed has been carried out, showing a good performance of the investigated OFLs. Back-calculated friction values were equal to, or even slightly lower than, reference O/W emulsions. The strip cleanliness with OFLs is much better than it is with the reference O/W emulsions. Only for a very thin product, as is the case in tinplate rolling, does the direct application of a conventional O/W dispersion (a high-particle-sized O/W emulsion) give a better performance than the investigated OFLs. Further development of OFLs should focus on this aspect.

In the context of electric steelworks, energy management is a key factor to reduce production costs while ensuring satisfaction of energy demands of all the different processes. Energy consumption optimization in the steel production chain can only be achieved by jointly considering individual processes as a network of users, in which each process is already close to the optimal operating point. This results in a large number of energy consumers to be managed, and to this aim a scheduler is usually adopted. The European project EnerMIND aims at effective and efficient steel production and a high utilization rate of production facilities through the implementation of a software demonstrator of a new energy management system based on a new energy management model exploiting a flexible infrastructure. This demonstrator, which covers the whole production chain, considers the areas of the steelworks showing the highest energy demands, i.e. Electric Arc Furnace, secondary steelmaking and the reheating furnaces that feed the rolling mills, in terms of total and peak values.