Singular Spectrum Analysis for Structural Damage Detection in an Electricity Pylon

Understanding the dynamic behavior of electricity pylons is essential for evaluating the integrity and resilience of their foundations, particularly under environmental loading and long-term degradation. Within the framework of the HEAT-COFFEE project, this study explores the use of Multi-channel Singular Spectrum Analysis (MSSA) as a non-invasive numerical tool for detecting structural damage in a finite element model of an electricity pylon. A coupled soil–structure model of the pylon was developed in Code_Aster, where the system was subjected to ambient noise excitation to reproduce realistic vibration conditions. The recorded responses at multiple points of the structure and foundation were processed using MSSA to extract dominant modes, study signal correlations, and reconstruct the dynamic response. Simulated damage was introduced by reducing the stiffness of selected elements, both in the pylon and in the surrounding soil, to evaluate the sensitivity of the method. The results show that, while the dominant frequencies remain nearly constant, subtle variations can be observed in the spectral content of the reconstructed components when structural damage is introduced. These variations are localized and depend on the measurement position along the structure. In contrast, the soil-damaged configuration remains largely indistinguishable from the healthy case. These findings indicate that MSSA can help capture subtle alterations in dynamic behavior caused by localized stiffness degradation. The study highlights the potential of MSSA as a vibration-based technique for non-invasive monitoring of large lattice structures and their foundations, providing a promising step toward ambient vibration–based damage detection in electricity pylons.