REWAS 2022: Automation and Digitalization for Advanced Manufacturing: On-Demand Oral Presentations
Sponsored by: TMS Extraction and Processing Division, TMS: Recycling and Environmental Technologies Committee, TMS: Process Technology and Modeling Committee
Program Organizers: Elsa Olivetti, Massachusetts Institute of Technology; Alexandra Anderson, Gopher Resource; Mertol Gokelma, Izmir Institute of Technology; Camille Fleuriault, Eramet Norway; Kaka Ma, Colorado State University

Monday 8:00 AM
March 14, 2022
Room: Energy & Environment (including REWAS 2022 Symposia)
Location: On-Demand Room


Refractory Lifetime Prediction in Industrial Processes with Artificial Intelligence: Nikolaus Voller1; Christoph Pichler1; Christine Wenzl1; Gregor Lammer1; 1RHI Magnesita
    This paper deals with refractory lifetime prediction by using artificial intelligence (AI). Through the effective use of process parameters, obtained from various operational processes within the Industrial (Cement/Lime, Non-Ferrous Metals, Process Industries, Foundry) and Steel sector, an AI model is generated. With the assistance of modern surveying technology, a correlation can be identified between process parameters and refractory wear. Using this method, suitable prediction of the refractory lifetime, as well as the wear mechanism, is possible. In addition, maintenance cycles can be adjusted and the optimal maintenance intensity of the operated furnaces can be ensured. With our intelligent Automated Process Optimization (APO) solution the prediction of refractory lifetime and wear mechanism can be done in real-time. Thus, we can provide value-added service to our customers.

Audio Signal Processing for Quantitative Moulding Material Regeneration: Philine Kerst1; Sebastian Tewes1; 1University of Duisburg-Essen
     As a natural product with limited resources, sand is of existential importance for various industries. Foundry technology in particular requires considerable quantities worldwide for sand moulds and cores. Depending on the process, the sand is recycled and thus reused. However, especially in the case of inorganically bound moulding sand, it is still frequently not recycled.Existing regeneration methods include mechanical, pneumatic, or combined processes. These have been developed, but have reached their analytical optimisation limits, as the processes are not transparent and make insitu analyses of moulding materials impossible. Within the scope of this research work, a methodology for the computer-aided processing of sound and image data was developed with the help of Convolutional Neural Networks (CNN), which is to evaluate the non-measurable changes of the moulding material in the running process in real time via process acoustics. The aim is to optimise process control in terms of time, cost, and energy efficiency.

Evolution of Process Models to Digital Twins: Alex Holtzapple1; 1Metsim International, LLC, USA
    Process simulation in the mining and metals industry was introduced in the early 1960s, with the invention of the computer, and has been rapidly evolving since that time. Programming languages and calculation methodologies have been constantly expanding to encompass innovative technology and equipment into metallurgical software for use by engineers and operators. Essentially no longer bound by data storage space or model convergence time, the next stage of this evolution is the incorporation of valuable process models into operations. “Digital twin” is the commonly accepted term for this, and certainly defines the overall simulation system accurately, though does not stress the importance of the well-calibrated, robust process models required. Constructing a multi-disciplined metallurgical, engineering and operational team ensures reliable results and recommendations from any operation’s digital twin.