Scholarlink

<p>The dynamic coupling of material and energy flows, compounded by delayed cross-process coordination, poses significant challenges to energy efficiency optimization in steel manufacturing. This study proposes a Dynamic Graph Convolution-Synergistic Fusion Model (DGCN-Synergy) to address these issues by integrating spatio-temporal graph learning with metallurgical process constraints. Key innovations include: (1) a dual-driven dynamic graph mechanism fusing data-driven features and metallurgical rules to capture real-time process correlations; (2) a three-level synergy framework optimizing micro-stability, meso-recovery, and macro-efficiency via closed-loop feedback; (3) adaptive spatio-temporal attention to resolve heterogeneous material-energy interactions. Industrial validation in a 6 Mt/year steel mill demonstrates that DGCN-Synergy outperforms conventional methods, achieving a 5.8% reduction in comprehensive energy consumption and 510,720 tCO₂/year emission reduction. These results highlight its potential as an interpretable and resilient solution for low-carbon transformation in complex industrial systems, providing a technical paradigm for energy-intensive industries pursuing carbon neutrality.</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

Dynamic graph convolution-driven synergy optimization of material and energy flows in iron and steel enterprises

  • 期刊论文

  • 发表日期：

    2025年12月20日

查看全文
  • Taian Qu,
  • Junsheng Wang,
  • Jing Zhang

摘要

The dynamic coupling of material and energy flows, compounded by delayed cross-process coordination, poses significant challenges to energy efficiency optimization in steel manufacturing. This study proposes a Dynamic Graph Convolution-Synergistic Fusion Model (DGCN-Synergy) to address these issues by integrating spatio-temporal graph learning with metallurgical process constraints. Key innovations include: (1) a dual-driven dynamic graph mechanism fusing data-driven features and metallurgical rules to capture real-time process correlations; (2) a three-level synergy framework optimizing micro-stability, meso-recovery, and macro-efficiency via closed-loop feedback; (3) adaptive spatio-temporal attention to resolve heterogeneous material-energy interactions. Industrial validation in a 6 Mt/year steel mill demonstrates that DGCN-Synergy outperforms conventional methods, achieving a 5.8% reduction in comprehensive energy consumption and 510,720 tCO₂/year emission reduction. These results highlight its potential as an interpretable and resilient solution for low-carbon transformation in complex industrial systems, providing a technical paradigm for energy-intensive industries pursuing carbon neutrality.