Scholarlink

<p>In the upcoming years, the steel industry will face major challenges due to the demands of CO<sub>2</sub> reduction and a&#xa0;shift towards circular economy. These developments lead to a&#xa0;larger diversity in crude steel compositions and thus to an increased throughput for secondary metallurgy including the Ruhrstahl Heraeus (RH) plant.</p><p>To meet these increased demands, the accuracy and level of detail of the process monitoring at the RH plant need to be improved. This contribution shows how this challenge might be met with a&#xa0;modelling approach based on computational fluid dynamics (CFD) combined with a&#xa0;data-driven method (recurrence CFD, rCFD) for the melt which will be linked to process data. This involves basics like measurement error estimation as well as sophisticated data and image evaluation. This contribution showcases the necessary steps to link an rCFD melt model to real plant data, thus working towards a&#xa0;digital twin of the RH process.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Making RH Fit for Green Steel Production: A Multi-method Approach to Process Monitoring

  • 期刊论文

  • 发表日期：

    2025年4月29日

查看全文
  • Maria Thumfart,
  • Xiaomeng Zhang,
  • Christine Gruber,
  • Johann Wachlmayr,
  • Stefan Pirker,
  • Roman Rössler

摘要

In the upcoming years, the steel industry will face major challenges due to the demands of CO2 reduction and a shift towards circular economy. These developments lead to a larger diversity in crude steel compositions and thus to an increased throughput for secondary metallurgy including the Ruhrstahl Heraeus (RH) plant.

To meet these increased demands, the accuracy and level of detail of the process monitoring at the RH plant need to be improved. This contribution shows how this challenge might be met with a modelling approach based on computational fluid dynamics (CFD) combined with a data-driven method (recurrence CFD, rCFD) for the melt which will be linked to process data. This involves basics like measurement error estimation as well as sophisticated data and image evaluation. This contribution showcases the necessary steps to link an rCFD melt model to real plant data, thus working towards a digital twin of the RH process.