Finite element modeling is central to the safety assessment and structural health monitoring of transmission towers, yet conventional workflows remain labor-intensive and depend on complete engineering drawings, which are often unavailable for aging towers. This study proposes an automated finite element modeling framework based on unmanned aerial vehicle (UAV) point clouds. Multi-view stereo reconstruction with geometric–photometric depth refinement is used to generate dense three-dimensional point clouds, from which key nodal coordinates are extracted and converted into executable ANSYS APDL command streams. Material properties, boundary conditions, and loads are then assigned automatically to construct computable finite element models. This method was first validated using concrete cubes and Q235B steel members and was further evaluated on a scaled transmission tower through comparison with the commercial SmartTower workflow. The results show that the proposed approach reduces modeling time from 1.5 days to 4.3 h, corresponding to an 88% efficiency improvement, while keeping nodal coordinate errors within 10% and modal frequency errors within 3.7%. The framework satisfies the basic requirements of transmission tower numerical analysis and provides a practical non-contact solution for the rapid assessment of aging towers lacking design documentation.
Jin et al. (Thu,) studied this question.