A 2.5D coupled FEM-BEM-MFS methodology for longitudinally invariant soil-structure interaction problems. (April 2021)

• Record Type: Journal Article
• Authors: Liravi, Hassan; Arcos, Robert; Ghangale, Dhananjay; Noori, Behshad; Romeu, Jordi
• Is Part Of: Computers and geotechnics. Volume 132(2021)
• Publication Date: 2021-04
• Subjects: Method of fundamental solutions -- 2.5D FEM-BEM-MFS -- Elastodynamics -- Soil-structure interaction -- Mesh-free methods -- Homogeneous full-space and half-space problems; Engineering geology -- Data processing -- Periodicals; Soil mechanics -- Data processing -- Periodicals; Rock mechanics -- Data processing -- Periodicals; 624.1510285
• DOI: 10.1016/j.compgeo.2021.104009
• Languages: English
• ISSNs: 0266-352X

Abstract

Highlights: A methodology to deal with soil-structure interaction problems is developed. A 2.5D FEM-BEM approach is used to model the structure and the surrounding soil. A 2.5D MFS approach is used to model the wave propagation through the soil. Accuracy and efficiency of the method are checked against a 2.5D FEM-BEM approach. A control technique is presented that minimises the MFS error. Abstract: This paper is concerned with a new prediction methodology for longitudinally invariant soil-structure interaction problems in elastodynamics. The method uses the finite-element method to model the structure, the boundary-element method to model the local soil surrounding the structure and the method of fundamental solutions to model the wave propagation through the soil, all of them formulated in the two-and-a-half dimensional domain. The proposed method firstly obtains the displacement field on the soil-structure interaction boundary making use of a two-and-a-half dimensional coupled finite element–boundary element method. The method of fundamental solutions is used then as a post-processing tool to compute the response of the soil, increasing the computational efficiency of the overall methodology with respect to a methodology that considers the boundary element method as a model of the wave propagation through the soil. The accuracy of the methodology is verified for four calculation examples: a solid cylinder and a circular thin shell embedded in a homogeneous full-space and also in a homogeneous half-space. This verification is performed comparing the results with available analytical or semi-analytical solutions and a conventional two-and-a-half dimensional coupled finite element–boundary element method. Furthermore, a control methodology to increase the robustness of the method is presented.