Scholarlink

<p>This paper presents novel numerical evaluations of the bearing capacity of conical foundations supported on dense sand. The study employs the Bolton yield criterion in conjunction with two-dimensional finite element limit analysis (FELA) methods under axisymmetric conditions. The parameter <i>b</i> is initially calibrated and estimated to be approximately 2.0, ensuring that the vertical ultimate capacity (<i>q</i><sub><i>u</i></sub>) predicted by the FELA aligns effectively with experimental data from previous research. The analysis of bearing capacity factors (<i>N</i><sub><i>γ</i></sub>) incorporates five key input parameters: the cone apex angle (<i>β</i>), particle crushing strength (<i>Q</i>), relative density (<i>D</i><sub><i>R</i></sub>), critical-state friction angle (<i>ϕ</i><sub><i>cv</i></sub>), and soil unit weight (<i>γ</i>). A comprehensive set of design charts is provided to facilitate detailed analysis and support decision-making in various design scenarios. Additionally, a novel soft-computing method is proposed for assessing the ultimate vertical load of conical footings in dense sand by integrating the FELA with the model known as the adaptive neuro-fuzzy inference system (ANFIS). The ANFIS model is further optimized using advanced computational techniques, including genetic algorithm (GA) and particle swarm optimization (PSO), to improve its accuracy, efficiency, and adaptability in handling complex nonlinear systems and predictive tasks. The hybrid GA-ANFIS and PSO-ANFIS models are validated and demonstrate strong agreement with the numerical results, with <i>R</i><sup>2</sup> values of 0.908 and 0.978 during training and 0.894 and 0.926 during testing, respectively. Comparative analysis reveals that the PSO–ANFIS model outperforms its counterparts, offering superior accuracy and robustness for predicting the bearing capacity of conical foundations in dense sand. These findings underscore the efficacy of hybrid soft computing techniques in geotechnical modeling and provide a reliable decision-support tool for engineering design and research applications. </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

Evaluating the Bearing Capacity of Conical Foundations on Dense Sand ~ FELA, Bolton Yield Criterion, and Hybrid ANFIS Algorithms

  • 期刊论文

  • 发表日期：

    2025年6月5日

查看全文
  • Nuchlee Boonjim,
  • Duy Tan Tran,
  • Wittaya Jitchaijaroen,
  • Suraparb Keawsawasvong,
  • Pitthaya Jamsawang

摘要

This paper presents novel numerical evaluations of the bearing capacity of conical foundations supported on dense sand. The study employs the Bolton yield criterion in conjunction with two-dimensional finite element limit analysis (FELA) methods under axisymmetric conditions. The parameter b is initially calibrated and estimated to be approximately 2.0, ensuring that the vertical ultimate capacity (qu) predicted by the FELA aligns effectively with experimental data from previous research. The analysis of bearing capacity factors (Nγ) incorporates five key input parameters: the cone apex angle (β), particle crushing strength (Q), relative density (DR), critical-state friction angle (ϕcv), and soil unit weight (γ). A comprehensive set of design charts is provided to facilitate detailed analysis and support decision-making in various design scenarios. Additionally, a novel soft-computing method is proposed for assessing the ultimate vertical load of conical footings in dense sand by integrating the FELA with the model known as the adaptive neuro-fuzzy inference system (ANFIS). The ANFIS model is further optimized using advanced computational techniques, including genetic algorithm (GA) and particle swarm optimization (PSO), to improve its accuracy, efficiency, and adaptability in handling complex nonlinear systems and predictive tasks. The hybrid GA-ANFIS and PSO-ANFIS models are validated and demonstrate strong agreement with the numerical results, with R2 values of 0.908 and 0.978 during training and 0.894 and 0.926 during testing, respectively. Comparative analysis reveals that the PSO–ANFIS model outperforms its counterparts, offering superior accuracy and robustness for predicting the bearing capacity of conical foundations in dense sand. These findings underscore the efficacy of hybrid soft computing techniques in geotechnical modeling and provide a reliable decision-support tool for engineering design and research applications.