Scholarlink

This paper will present an investigation into the performance of ANNs using data generated through Finite Element Analysis in order to model the relationship between excavation depth and influential factors such as Horizontal Displacement, Shear Forces, Bending Moment, and Axial Forces. The aforementioned factors are the main contributors to the excitatory controlling of excavations. The ANN approach is resorted to because it is a more capable tool to analyze and predict nonlinear complex systems. These FEA data are Horizontal Displacement, Shear Forces, Bending Moment, and Axial Forces values, which form the input for the ANN model. The target output variable is Excavation Depth. The estimated results of the study show that the performance of ANN, done by a value of R2 equal to 0.959955, proves very high accuracy in the prediction of excavation depth based on the input factors taken from FEA. To name a few, Axial Forces had the most influence on the depth of excavation, whereas Shear Forces had a lesser effect. ANN is an authenticated tool in various ways within the field of prediction and optimization of excavation design and improvement of accuracy in construction projects, contributing to both safety and cost savings during the construction process. Incorporation of FEA data into the ANN model also denotes a new directionality in the application of the artificial intelligence techniques in geotechnical analysis.

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 Impact of Horizontal Displacement and Forces on Excavation Depth Using Artificial Neural Networks and Finite Element Analysis

  • 章节

  • 发表日期：

    2025年

查看全文
  • Truong Xuan Dang,
  • Tuan Anh Nguyen,
  • Hoa Van Vu Tran

摘要

This paper will present an investigation into the performance of ANNs using data generated through Finite Element Analysis in order to model the relationship between excavation depth and influential factors such as Horizontal Displacement, Shear Forces, Bending Moment, and Axial Forces. The aforementioned factors are the main contributors to the excitatory controlling of excavations. The ANN approach is resorted to because it is a more capable tool to analyze and predict nonlinear complex systems. These FEA data are Horizontal Displacement, Shear Forces, Bending Moment, and Axial Forces values, which form the input for the ANN model. The target output variable is Excavation Depth. The estimated results of the study show that the performance of ANN, done by a value of R2 equal to 0.959955, proves very high accuracy in the prediction of excavation depth based on the input factors taken from FEA. To name a few, Axial Forces had the most influence on the depth of excavation, whereas Shear Forces had a lesser effect. ANN is an authenticated tool in various ways within the field of prediction and optimization of excavation design and improvement of accuracy in construction projects, contributing to both safety and cost savings during the construction process. Incorporation of FEA data into the ANN model also denotes a new directionality in the application of the artificial intelligence techniques in geotechnical analysis.