Scholarlink

<p>Sensor-based sorting (SBS) is increasingly vital for upgrading low-grade and geologically complex mineral deposits, yet the amenability of individual fragments to SBS is still largely assessed through trial-and-error testing. Empirical studies consistently demonstrate that a fragment’s sortability is strongly governed by its mineralogical homogeneity. Across diverse scientific fields, including chemistry, materials science, and medical diagnostics, quantifying homogeneity has played a central role in technological advancement. Methods have evolved from early statistical measures in the 1700s to sophisticated modern image-analysis techniques. In geoscience, the need to quantify spatial grade variability has long been recognized, leading to the application of tools such as kriging, Global Moran’s I, entropy-based measures, and Gy’s theories of distribution heterogeneity. This paper highlights the need to develop a homogeneity parameter tailored to the characterization of rock fragments intended for sensor-based particle sorting. These fragments, typically in the centimeter (meso) size range, are sorted as early as possible in the mining value chain. This study reviews a broad range of statistical and analytical approaches used across scientific disciplines to quantify variance and homogeneity in materials, populations, and images. Each method is evaluated for its relevance and applicability to mineral fragments, with attention to its strengths and limitations. Through this evaluation, the key parameters necessary for constructing an effective homogeneity metric for rock fragments are identified, establishing a foundation for the development of a metric that can characterize SBS feed and support rigorous assessment of sorter performance.</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

A Review of Homogeneity Measurements and their Importance to Advancing Sensor-based Ore Sorting

  • 期刊论文

  • 发表日期：

    2026年7月10日

查看全文
  • Jane Danoczi,
  • Christopher Hawkes,
  • Laura Smith

摘要

Sensor-based sorting (SBS) is increasingly vital for upgrading low-grade and geologically complex mineral deposits, yet the amenability of individual fragments to SBS is still largely assessed through trial-and-error testing. Empirical studies consistently demonstrate that a fragment’s sortability is strongly governed by its mineralogical homogeneity. Across diverse scientific fields, including chemistry, materials science, and medical diagnostics, quantifying homogeneity has played a central role in technological advancement. Methods have evolved from early statistical measures in the 1700s to sophisticated modern image-analysis techniques. In geoscience, the need to quantify spatial grade variability has long been recognized, leading to the application of tools such as kriging, Global Moran’s I, entropy-based measures, and Gy’s theories of distribution heterogeneity. This paper highlights the need to develop a homogeneity parameter tailored to the characterization of rock fragments intended for sensor-based particle sorting. These fragments, typically in the centimeter (meso) size range, are sorted as early as possible in the mining value chain. This study reviews a broad range of statistical and analytical approaches used across scientific disciplines to quantify variance and homogeneity in materials, populations, and images. Each method is evaluated for its relevance and applicability to mineral fragments, with attention to its strengths and limitations. Through this evaluation, the key parameters necessary for constructing an effective homogeneity metric for rock fragments are identified, establishing a foundation for the development of a metric that can characterize SBS feed and support rigorous assessment of sorter performance.