A Sensor Fault Detection Algorithm for Structural Health Monitoring Using Adaptive Differential Evolution

New sensor technologies make it possible to monitor structures with a dense array of sensors. In a dense sensor network with low-cost sensor nodes, it is quite probable that some nodes fail, resulting in unreliable data for inference and decision. In a large sensor network, the computational time in isolating the faulty sensor using the current algorithms is prohibitive for online sensor fault isolation. In this article, we combine the PCA-based algorithm with an adaptive differential evolution algorithm to improve the performance of sensor fault isolation. Numerical investigations are presented to demonstrate the effectiveness of the proposed algorithm.