Scholarlink

<p>The global rise in lithium demand, driven by the expansion of the new energy sector, underscores the need for efficient mineral processing of spodumene, the primary lithium ore. This study addresses key challenges in current pre-sorting systems, including low sorting precision, high energy consumption, and limited real-time adaptability, by developing an intelligent X-ray transmission (XRT) pre-sorting method enhanced with a novel Mineral-YOLOv8s deep learning algorithm. The methodology integrates machine vision and artificial intelligence to achieve lightweight, accurate ore detection under real-world mining conditions, leveraging innovative neural network improvement. Key findings demonstrate that the Mineral-YOLOv8s algorithm enhances sorting precision by almost 10%, reduces computational complexity by almost 51%, and improves detection speed by 25.4 frames per second compared to baseline models. Industrial tests confirm its effectiveness across varying ore particle sizes, processing volumes, and conveyor speeds. This technology significantly improves resource utilization, reduces environmental impacts, and optimizes efficiency in spodumene beneficiation. The proposed framework offers a scalable solution for advancing low-carbon, sustainable practices in mineral processing.</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

Intelligent Pre-sorting Method for Spodumene Ore Combining X-ray Transmission (XRT) and Deep Learning: Promoting Energy Conservation and Emission Reduction in Mines

  • 期刊论文

  • 发表日期：

    2025年7月26日

查看全文
  • Tianyou Yu,
  • Yimin Zhu,
  • Jie Liu,
  • Yuexin Han,
  • Yanjun Li

摘要

The global rise in lithium demand, driven by the expansion of the new energy sector, underscores the need for efficient mineral processing of spodumene, the primary lithium ore. This study addresses key challenges in current pre-sorting systems, including low sorting precision, high energy consumption, and limited real-time adaptability, by developing an intelligent X-ray transmission (XRT) pre-sorting method enhanced with a novel Mineral-YOLOv8s deep learning algorithm. The methodology integrates machine vision and artificial intelligence to achieve lightweight, accurate ore detection under real-world mining conditions, leveraging innovative neural network improvement. Key findings demonstrate that the Mineral-YOLOv8s algorithm enhances sorting precision by almost 10%, reduces computational complexity by almost 51%, and improves detection speed by 25.4 frames per second compared to baseline models. Industrial tests confirm its effectiveness across varying ore particle sizes, processing volumes, and conveyor speeds. This technology significantly improves resource utilization, reduces environmental impacts, and optimizes efficiency in spodumene beneficiation. The proposed framework offers a scalable solution for advancing low-carbon, sustainable practices in mineral processing.