Sensor fault detection in structural health monitoring using null subspace–based approach

Abstract

It is essential to have a robust sensor fault detection and isolation algorithm integrated with successful online continuous structural health monitoring scheme to avoid false alarms. In this work, a sensor fault detection and isolation technique based on null subspace method is presented. The robustness of the proposed sensor failure detection and isolation algorithm is demonstrated using the vibration data obtained from an experimental study of a scaled down bridge model. Studies presented in this article clearly indicate that the proposed method is robust in identification of exact time instant of sensor fault and also isolation of faulty sensor. The proposed algorithm is equally good for all possible types of sensor faults (both additive and multiplicative) and also capable of isolating multiple sensor faults in spatial location of the structure.