Modeling of tectonic faults using finite stiffness links with integration in CAE Fidesys

This paper presents an original methodology for modeling tectonic faults in rock masses using finite stiffness links, developed within the framework of the finite element method. Unlike traditional approaches that require explicit construction of thefault geometry and its integration with theother structural elements of the model, the proposed method allows tectonic faults to be defined implicitly. This is achieved through the introduction of special spring elements between themesh nodes, which have adjustable stiffness in thespecified directions. This approach to representing weakened zones eliminates the need to modify the base geometry of the model, simplifies its topology, and improves the stability of numerical calculations. The methodology is implemented as a standalone Python-based module and is used in conjunction with the domestic CAE Fidesys software package. The module automatically generates a list of finite stiffness links based on the input coordinates of thefaults, assignsthe stiffness parameters, and integrates them into the computational model without modifying the graphical interface of the main program. The developed tool has been tested on a number of geomechanical problems and it has beensuccessfully verified through comparisons with thetheoretical and empirical results. In addition, the article provides recommendations for selecting thespring element parameters based on the geological structure of the rock mass, the degree of tectonic fracturing, and the specific features of the engineering problem. The developed methodology can be efficiently applied in designing of mining excavations, engineering assessment of rock stability in seismically active regions, and geomechanical support of miningprojects. Its use is particularly relevant in conditions of complex geological structures and multiple intersecting faults, where classical modeling methods face significant computational and methodological limitations. © 2025 Elsevier B.V., All rights reserved.