1,721,093 research outputs found
A big step ahead in Metal Science and Technology through the application of Artificial Intelligence
No full textDigitalization is a key enabling factor of the dramatic transformation of metal industry since the 90-ies. Important applications were implemented since then, exploiting multi-physical modelling, complex real time process control and Machine Learning. This trend was further enhanced and accelerated by Industry 4.0. Digitalization implies harvesting impressive volumes of heterogeneous data, which need to be stored, processed and, mostly, interpreted to extract relevant information and “knowledge”. Knowledge means capability of interpreting data, of explaining and representing material transformation and product evolution during the different process stages considering complex interactions among process and product variables, including aspects still not perfectly understood. Artificial Intelligence supports knowledge extraction by enabling optimal process management and control, higher flexibility and product quality, stronger resource and energy efficiency, namely sustainability. Challenges and opportunities of enhancing metallurgical science and technology through Artificial intelligence are considered in this review
Automatic steel grades design for Jominy profile achievement through neural networks and genetic algorithms
No full textThe paper proposes an approach to the design of the chemical composition of steel, which is based on neural networks and genetic algorithms and aims at achieving a desired hardenability behavior possibly matching other constraints related to the steel production. Hardenability is a mechanical feature of steel, which is extremely relevant for a wide range of steel applications and refers to the steel capability to improve its hardness following a heat treatment. In the proposed approach, a neural-network-based predictor of the so-called Jominy hardenability profile is exploited, and an optimization problem is formulated, where the optimization function allows taking into account both the desired accuracy in meeting the target Jominy profile and other constraint. The optimization is performed through genetic algorithms. Numerical results are presented and discussed, showing the efficiency of the proposed approach together with its flexibility and easy customization with respect to the user demands and production objectives
A Novel Approach to Jominy Profile Prediction Based on 1D Convolutional Neural Networks and Autoencoders that Supports Transfer Learning
Full text linkEditorial for the Special Issue: Overview, state of the art, recent developments and future trends regarding Hydrogen route for a green steel making process
No full text
A Combined Approach for Enhancing the Stability of the Variable Selection Stage in Binary Classification Tasks
No full textVariable selection is an essential tool for gaining knowledge on a problem or phenomenon, by identifying the factors that shows the highest influence on it. It is also fundamental for the implementation of machine learning-based approaches to modelling and classification tasks, by improving performances and reducing computational cost. Furthermore, in many real-world applications, such as the ones in the medical field, a relevant number of variables are jointly observed, but the number of available observations is quite limited. In these cases, variable selection is clearly essential, but standard variable selection approaches become “unstable”, as the high correlation among different variables or their similar relevance with respect to the considered target lead to multiple solutions leading to similar performances. In machine-learning based classification, the stability of variable selection, namely its robustness with respect variations in the classifier training dataset, is as important as the performance of the classifier itself. The paper presents an automatic procedure for variable selection in classification tasks, which ensures excellent stability of the selection and does not require any a priori information on the available data
A way to reduce environmental impact of ladle furnace slag
No full textIn the last decades the European steel industry has made continuous efforts to reduce residues
and byproducts and to increase recycling in order to reduce its environmental impact. While
some steelmaking slags have been widely characterised and, to a certain extent reused, ladle
furnace (LF) slag is used in different applications because of its specific properties. The main
purpose of the case study presented in this paper concerns the reduction of potential LF slag
environmental impacts, because of its intrinsic physicochemical properties. During the handling
and cooling of LF slag, it disintegrates into a powder due to instability of the dicalcium silicate,
causing an increase in dust emissions to the environment. The approach presented in this paper
aims to reduce this phenomenon in order to achieve a more sustainable solution in term of
reduction of powder dispersion in the environment, of costs saving and of nuisance reduction in
the surroundings areas
High-strength steels for the automotive sector: A simple austenite transformation model for continuous annealing and galvanizing lines
No full textIn continuous annealing and hot-dip galvanizing processes, the final steel structures heavily depend on the process parameters (e.g. times and temperatures), which, thus, must be accurately tuned. In particular, the short-term permanence in a two-phase ferrite + austenite field and the following cooling process have a strong impact and can be very articulated, according to the plant type. Therefore, it is extremely useful to have practical and simple models estimating the austenite transformation based on the undergone thermal cycle. Actually, the process speed, especially for the cooling stage, implies that standard CCT curves are not applicable for determining the final structure. This work presents a solution exploiting a simple metallurgical-mathematical model. Such model, on the basis of experimental tests and starting from the chemical composition and the thermal cycle, estimates both the final structure and the mechanical properties of Advanced High Strength Steels, which are of utmost interest in the automotive sector
Application of Hydrogen Permeation Techniques to Assess the Effect of Inhibitors in Hydrochloric Steel Pickling of Low-C Steels and Interstitial-Free Steels
No full textThe acid pickling phase, generally entrusted to HCl-based solutions, plays a fundamental role in the production of low-C and Interstitial Free steel coils. Very often, however, pickling operations are challenging due to the possible onset of surface defects (under-pickling, over-pickling, appearance of surface defects, etc.) linked to complex phenomena involving numerous factors, such as acid concentration, Fe ions in solution, temperature, time, steel type and scale structure. Moreover, the use of inhibitors, which are essential for controlling the pickling process, represents one of the major challenges. Their dosage, however minimal, must be carefully chosen based on type of processed coils and other plant engineering variables. The flow of hydrogen that is generated on the surface of the steel during etching causes absorption of hydrogen by the metal and its measurement is strictly linked to the phenomena occurring during scale removal. The correlation of data obtained from a Devanahan double-cell hydrogen permeation system to results of other techniques, such as weight loss, surface analysis, metallography, electron microscopy show that this system can be effectively used to indirectly measure the solution aggressiveness. This experimental approach is applied to two industrial steels: a simple carbon base and an Interstitial Free Ti steel
- …
