Sensor optimization for progressive damage diagnosis in complex structures

We propose a sequential Monte Carlo (SMC) based progressive structural damage diagnosis framework that tracks damage by integrating information from physics-based damage evolution models and using stochastic relationships between the measurements and the damage. The approach described in this paper adaptively configures the sensors used to collect the measurements using the minimum predicted mean squared error (MSE) as the performance metric. Optimization is performed globally over the entire search space of all available sensors. Results are presented for the diagnosis of fatigue damage in a notched laminate, demonstrating the effectiveness of the proposed method.