Scholarlink

The scrap-basedScrap recycling electric arc furnaceElectric arc furnace (EAF) is pivotal in sustainable steelmakingSteelmaking by recyclingRecycling steel and minimizing raw materialRaw materials extraction. The precise control of phosphorus, a critical impurity affecting steel quality, remains a significant challenge in the industryIndustry. This work details the development of an advanced artificial neural network (ANNArtificial Neural Network (ANN)) model designed to predict the final phosphorus content of steel based on the process parameters within an EAF. This model leverages systematic data integration and rigorous model validation, demonstrating superior predictive accuracy compared to existing models. Inherent model limitations will also be addressed and future research directions aimed at further enhancing predictive capabilities and expanding the applicability of the proposed approach in steelmakingSteelmaking context will be presented. Industrial implementation of the model will be discussed, highlighting opportunities to optimize EAF operations for improved green steel quality.

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

Controlling Minor Element Phosphorus in Green Electric Steelmaking Using Neural Networks

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Elmira Moosavi-Khoonsari,
  • Riadh Azzaz,
  • Valentin Hurel,
  • Mohammad Jahazi,
  • Samira Ebrahimi Kahou

摘要

The scrap-basedScrap recycling electric arc furnaceElectric arc furnace (EAF) is pivotal in sustainable steelmakingSteelmaking by recyclingRecycling steel and minimizing raw materialRaw materials extraction. The precise control of phosphorus, a critical impurity affecting steel quality, remains a significant challenge in the industryIndustry. This work details the development of an advanced artificial neural network (ANNArtificial Neural Network (ANN)) model designed to predict the final phosphorus content of steel based on the process parameters within an EAF. This model leverages systematic data integration and rigorous model validation, demonstrating superior predictive accuracy compared to existing models. Inherent model limitations will also be addressed and future research directions aimed at further enhancing predictive capabilities and expanding the applicability of the proposed approach in steelmakingSteelmaking context will be presented. Industrial implementation of the model will be discussed, highlighting opportunities to optimize EAF operations for improved green steel quality.