Scholarlink

<p>This study tackles key challenges in seabed sediment acoustic modeling-parameter optimization and acoustic velocity dispersion prediction-by introducing an integrated framework combining global sensitivity analysis with multi-objective optimization. The approach systematically uncovers the physical mechanisms of influential parameters and delivers a high‑precision acoustic prediction model for typical South China Sea sediments. Results show that porosity, as the fundamental porous‑media parameter, maintains the highest sensitivity (index &gt; 0.45) across the full 10&#xa0;Hz–100&#xa0;kHz range. Parameter sensitivity displays strong frequency dependence: the Material exponent in the Buckingham‑GS model grows exponentially above 1&#xa0;kHz, underscoring the dominance of grain‑contact mechanisms, whereas frame parameters in Biot‑Stoll‑type models gain marked sensitivity above 10&#xa0;kHz, reflecting frequency‑dependent poroelastic coupling. Using a multi‑objective genetic algorithm, sound speed (deviation &lt; ± 12&#xa0;m/s) and attenuation (deviation &lt; ± 20 dB/m) were co‑optimized for three South China Sea sediments (silty sand, silt, silty clay). Clear model‑application ranges emerge: BICSQS is most robust for broadband prediction; Buckingham‑VGS suits low‑frequency scenarios; and Biot‑Stoll‑type models better predict high‑frequency attenuation in coarse‑grained sediments. The developed “mechanism‑parameter‑selection” framework offers a reliable parametric foundation for marine geoacoustic modeling, with practical relevance for resource exploration and sonar‑system optimization.</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

Modeling the acoustic properties of South China sea sediments: parameter sensitivity analysis and multi-objective optimization approach

  • 期刊论文

  • 发表日期：

    2026年3月2日

查看全文
  • Qingjie Zhou,
  • Jingqiang Wang,
  • Guangming Kan,
  • Danping Cao

摘要

This study tackles key challenges in seabed sediment acoustic modeling-parameter optimization and acoustic velocity dispersion prediction-by introducing an integrated framework combining global sensitivity analysis with multi-objective optimization. The approach systematically uncovers the physical mechanisms of influential parameters and delivers a high‑precision acoustic prediction model for typical South China Sea sediments. Results show that porosity, as the fundamental porous‑media parameter, maintains the highest sensitivity (index > 0.45) across the full 10 Hz–100 kHz range. Parameter sensitivity displays strong frequency dependence: the Material exponent in the Buckingham‑GS model grows exponentially above 1 kHz, underscoring the dominance of grain‑contact mechanisms, whereas frame parameters in Biot‑Stoll‑type models gain marked sensitivity above 10 kHz, reflecting frequency‑dependent poroelastic coupling. Using a multi‑objective genetic algorithm, sound speed (deviation < ± 12 m/s) and attenuation (deviation < ± 20 dB/m) were co‑optimized for three South China Sea sediments (silty sand, silt, silty clay). Clear model‑application ranges emerge: BICSQS is most robust for broadband prediction; Buckingham‑VGS suits low‑frequency scenarios; and Biot‑Stoll‑type models better predict high‑frequency attenuation in coarse‑grained sediments. The developed “mechanism‑parameter‑selection” framework offers a reliable parametric foundation for marine geoacoustic modeling, with practical relevance for resource exploration and sonar‑system optimization.