Scholarlink

<p>This study explores the application and comparative performance of the Differential Evolution (DE) algorithm and the Genetic Algorithm (GA) for interpreting magnetic anomalies caused by 2D dipping dyke-like structures, emphasizing their utility for both mineral exploration and engineering geological investigations. Accurate estimations of dyke parameters (depth, width, dip, magnetization) is critical for: (1) assessing geotechnical stability in construction and tunneling projects; (2) evaluating volcanic and seismic hazards related to dyke propagation; and (3) identifying mineralized zones that impact groundwater flow. The dyke parameters are estimated by minimizing an objective function to achieve an optimal match with observed data. The performance of the DE approach was tested on synthetic magnetic datasets under various noise conditions (0%, 5%, 10%, 15%, and 20%). The approach was further validated using real-world magnetic datasets from Gansu Province, China. A comparative analysis with the Genetic Algorithm (GA) demonstrates that DE offers superior efficiency and reliability, particularly in high-noise scenarios. The inversion results align well with geological and geophysical constraints derived from borehole data, confirming the effectiveness of DE for subsurface characterization. This approach enhances site investigation accuracy and infrastructure planning, offering a robust tool for geotechnical and environmental assessments.</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

Comparative study of differential evolution (DE) and genetic algorithm (GA) for robust magnetic inversion and subsurface characterization of 2D dipping dykes in engineering geological investigations

  • 期刊论文

  • 发表日期：

    2026年3月9日

查看全文
  • Reza Toushmalani,
  • Yves Géraud,
  • Khalid S. Essa

摘要

This study explores the application and comparative performance of the Differential Evolution (DE) algorithm and the Genetic Algorithm (GA) for interpreting magnetic anomalies caused by 2D dipping dyke-like structures, emphasizing their utility for both mineral exploration and engineering geological investigations. Accurate estimations of dyke parameters (depth, width, dip, magnetization) is critical for: (1) assessing geotechnical stability in construction and tunneling projects; (2) evaluating volcanic and seismic hazards related to dyke propagation; and (3) identifying mineralized zones that impact groundwater flow. The dyke parameters are estimated by minimizing an objective function to achieve an optimal match with observed data. The performance of the DE approach was tested on synthetic magnetic datasets under various noise conditions (0%, 5%, 10%, 15%, and 20%). The approach was further validated using real-world magnetic datasets from Gansu Province, China. A comparative analysis with the Genetic Algorithm (GA) demonstrates that DE offers superior efficiency and reliability, particularly in high-noise scenarios. The inversion results align well with geological and geophysical constraints derived from borehole data, confirming the effectiveness of DE for subsurface characterization. This approach enhances site investigation accuracy and infrastructure planning, offering a robust tool for geotechnical and environmental assessments.