Scholarlink

<p>This study presents a novel integrated approach that combines four machine learning (ML) based equations for predicting the swelling–consolidation envelope of expansive soils, from which the swelling index, swelling pressure, swelling potential, and compression index can be derived. The four ML based equations were originally developed for independently estimating the various swell and compression properties, thereby alleviating the need for time-consuming and expensive oedometer tests. The proposed integrated approach was applied to six geographically diverse expansive soils from Canada, the United States, Saudi Arabia, and India, and the predicted envelopes were compared with laboratory measurements conducted in accordance with ASTM D4546 and ASTM D2435. There is a good comparison between the predicted and measured swelling-consolidation envelopes, indicating that the equations are compatible when they are used in an integrated manner. Any or all key parameters required can be derived from the predicted envelope for use in preliminary assessments required in geotechnical engineering practice and for planning ground improvement methods for addressing challenges associated with expansive soils.</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

Validation of Machine Learning Models for Predicting the Swelling–Consolidation Envelope of Expansive Soils

  • 期刊论文

  • 发表日期：

    2026年3月13日

查看全文
  • Aolin Zhang,
  • Sai K. Vanapalli

摘要

This study presents a novel integrated approach that combines four machine learning (ML) based equations for predicting the swelling–consolidation envelope of expansive soils, from which the swelling index, swelling pressure, swelling potential, and compression index can be derived. The four ML based equations were originally developed for independently estimating the various swell and compression properties, thereby alleviating the need for time-consuming and expensive oedometer tests. The proposed integrated approach was applied to six geographically diverse expansive soils from Canada, the United States, Saudi Arabia, and India, and the predicted envelopes were compared with laboratory measurements conducted in accordance with ASTM D4546 and ASTM D2435. There is a good comparison between the predicted and measured swelling-consolidation envelopes, indicating that the equations are compatible when they are used in an integrated manner. Any or all key parameters required can be derived from the predicted envelope for use in preliminary assessments required in geotechnical engineering practice and for planning ground improvement methods for addressing challenges associated with expansive soils.