Damage-adaptive Optimization of Sensor Number and Placement for Structural Health Monitoring Using Transmissibility Functions
Please login to view abstract download link
Efficient sensor deployment is essential in Structural Health Monitoring (SHM) to ensure accurate detection and localization of structural damage. While most optimization methods focus solely on sensor placement to capture the dynamic behavior of an undamaged structure, this study proposes a damage-adaptive approach that simultaneously optimizes both the number and placement of sensors across multiple damage scenarios. The method evaluates several optimized sensor configurations, each tailored to a specific damage case, and derives a minimal yet highly sensitive sensor layout using a transmissibility-based damage indicator. For each scenario, an optimal sensor set is identified based on transmissibility deviations. These configurations are then consolidated into a sensor occurrence matrix, and a unified sensor layout is extracted using computed sensor importance scores. The approach is demonstrated on a multi-storey reinforced concrete building. Potential damage locations are selected by identifying structural elements (columns, beams, and shear walls) subjected to the highest internal forces under typical loadings. Damage is simulated by locally reducing stiffness in selected components. To assess the robustness of the unified configuration, its performance is evaluated across all damage scenarios using the Modal Assurance Criterion (MAC), a damage sensitivity index, and identification accuracy. Results indicate that the proposed method provides reliable damage sensitivity using a reduced number of strategically placed sensors. This approach is especially effective for structures where potential damage zones can be reasonably anticipated, offering a practical and scalable solution for damage-oriented SHM system design in complex buildings.