Heterogeneous Sensor Configuration Optimization in Dynamic Structural Response Estimation

Despite the availability of a wide range of sensors for Structural Health Monitoring (SHM) of civil engineering structures, practical constraints such as budget limitations and site accessibility often restrict the number of sensors that can be deployed. To overcome this limitation, it is imperative to strategically optimize sensor configurations to ensure accurate estimation of structural responses at arbitrary locations. This study presents a robust methodology for optimizing heterogeneous sensor configurations, effectively addressing model and measurement uncertainties. The approach integrates an Augmented Kalman Filter (AKF) for precise estimation of structural responses and a Genetic Algorithm (GA) for Sensor Configuration Optimization (SCO), aiming to minimize the average strain estimation error across all degrees of freedom. Critical questions concerning the optimal number, placement, and type of sensors are systematically investigated. The proposed method is validated through two comprehensive numerical case studies, a plate and a truss structure, demonstrating its high accuracy and reliability. The results confirm the superior performance of the proposed approach in enabling effective SHM through optimized heterogeneous sensor deployment.