Scholarlink

<p>This study investigates the application of a hybrid neural network model for delineating sulfide mineralization in complex geological settings. By integrating Artificial Neural Networks (ANN) with metaheuristic algorithms, specifically Artificial Bee Colony (ABC) and Particle Swarm Optimization (PSO), the research optimizes resistivity inversion techniques to enhance geophysical exploration. Field data were collected from the Nash Creek region in New Brunswick, Canada, utilizing electromagnetic (EM) surveys and Direct Current/Induced Polarization (DC/IP) methods across six survey lines configured with a Wenner array and 10-m electrode spacing. The results demonstrate that the hybrid model significantly outperforms traditional 2D resistivity inversion methods, effectively mapping mineralization zones characterized by low resistivity (&lt; 50&#xa0;Ω·m) and high chargeability Because the region is characterized by a significant glacial overburden and minimal outcrop exposure, verification was conducted using borehole drilling and core sampling within the western part of the study area. Notably, pyrite alteration zones were identified in regions with resistivity below 50&#xa0;Ω·m, exhibiting significant chargeability. Strong correlations between Versatile Time-Domain Electromagnetic (VTEM) survey results and IP data were observed, with Transient Electromagnetic (TEM) Reduced to Pole (RTP) data aligning closely with resistivity measurements. This study underscores a significant inverse correlation between resistivity and chargeability, effectively identifying pyrite-rich alteration zones. These findings validate the efficacy of the hybrid ANN-ABC-PSO model for geophysical exploration in challenging environments, offering valuable insights into mineral deposit structures and informing future exploration strategies. While shallow alteration zones were successfully mapped, further drilling is essential to confirm the extent and characteristics of the identified mineralization.</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

Efficient base metal exploration in northern New Brunswick, Canada through a hybrid ANN integrated with ABC and PSO methods

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年4月10日

查看全文
  • Claire Dusabemariya,
  • Wei Qian,
  • Romuald Bagaragaza,
  • Ajibola Richard Faruwa,
  • Ali Mossad

摘要

This study investigates the application of a hybrid neural network model for delineating sulfide mineralization in complex geological settings. By integrating Artificial Neural Networks (ANN) with metaheuristic algorithms, specifically Artificial Bee Colony (ABC) and Particle Swarm Optimization (PSO), the research optimizes resistivity inversion techniques to enhance geophysical exploration. Field data were collected from the Nash Creek region in New Brunswick, Canada, utilizing electromagnetic (EM) surveys and Direct Current/Induced Polarization (DC/IP) methods across six survey lines configured with a Wenner array and 10-m electrode spacing. The results demonstrate that the hybrid model significantly outperforms traditional 2D resistivity inversion methods, effectively mapping mineralization zones characterized by low resistivity (< 50 Ω·m) and high chargeability Because the region is characterized by a significant glacial overburden and minimal outcrop exposure, verification was conducted using borehole drilling and core sampling within the western part of the study area. Notably, pyrite alteration zones were identified in regions with resistivity below 50 Ω·m, exhibiting significant chargeability. Strong correlations between Versatile Time-Domain Electromagnetic (VTEM) survey results and IP data were observed, with Transient Electromagnetic (TEM) Reduced to Pole (RTP) data aligning closely with resistivity measurements. This study underscores a significant inverse correlation between resistivity and chargeability, effectively identifying pyrite-rich alteration zones. These findings validate the efficacy of the hybrid ANN-ABC-PSO model for geophysical exploration in challenging environments, offering valuable insights into mineral deposit structures and informing future exploration strategies. While shallow alteration zones were successfully mapped, further drilling is essential to confirm the extent and characteristics of the identified mineralization.