Scholarlink

<p>Active stability failure, typically caused by soil self-weight and surface surcharge, is a widely studied phenomenon. However, passive stability failure, where external loads oppose the self-weight of the soil, referred to as blowout, has not been extensively investigated. This paper explores the stability of three-dimensional rectangular blowout trapdoors in cohesive-frictional soils using three-dimensional finite element limit analysis (3D FELA). Stability factors, including the cohesion factor (<i>F</i><sub><i>c</i></sub>), surcharge factor (<i>F</i><sub><i>s</i></sub>), and unit weight factor (<i>F</i><sub><i>r</i></sub>), are determined using Terzaghi’s superposition method. Additionally, the dataset is analysed using Adaptive Neuro-Fuzzy Inference System (ANFIS) models, which are optimized through two advanced meta-heuristic techniques: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The results show that the ANFIS-PSO model outperforms the ANFIS-GA model. Specifically, the ANFIS-PSO model achieves coefficients of determination (<i>R</i><sup><i>2</i></sup>) of 0.962, 0.892, and 0.837 for <i>F</i><sub><i>c</i></sub>, <i>F</i><sub><i>s</i></sub>, and <i>F</i><sub><i>r</i></sub>, respectively, on the training dataset and 0.953, 0.898, and 0.629 for <i>F</i><sub><i>c</i></sub>, <i>F</i><sub><i>s</i></sub>, and <i>F</i><sub><i>r</i></sub>, respectively, on the testing dataset. The findings provide insights into the stability characteristics of blowout trapdoors, offering valuable data for the design of underground structures and safety assessments in geotechnical engineering.</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

Prediction of Blowout Stability in Underground Storage Tanks Using a Hybrid ANFIS–PSO Framework

  • 期刊论文

  • 发表日期：

    2025年12月19日

查看全文
  • Rithy Domphoeun,
  • Jim Shiau,
  • Suraparb Keawsawasvong,
  • Pitthaya Jamsawang

摘要

Active stability failure, typically caused by soil self-weight and surface surcharge, is a widely studied phenomenon. However, passive stability failure, where external loads oppose the self-weight of the soil, referred to as blowout, has not been extensively investigated. This paper explores the stability of three-dimensional rectangular blowout trapdoors in cohesive-frictional soils using three-dimensional finite element limit analysis (3D FELA). Stability factors, including the cohesion factor (Fc), surcharge factor (Fs), and unit weight factor (Fr), are determined using Terzaghi’s superposition method. Additionally, the dataset is analysed using Adaptive Neuro-Fuzzy Inference System (ANFIS) models, which are optimized through two advanced meta-heuristic techniques: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The results show that the ANFIS-PSO model outperforms the ANFIS-GA model. Specifically, the ANFIS-PSO model achieves coefficients of determination (R2) of 0.962, 0.892, and 0.837 for Fc, Fs, and Fr, respectively, on the training dataset and 0.953, 0.898, and 0.629 for Fc, Fs, and Fr, respectively, on the testing dataset. The findings provide insights into the stability characteristics of blowout trapdoors, offering valuable data for the design of underground structures and safety assessments in geotechnical engineering.