Structural health monitoring of steel frames using a network of self-powered strain and acceleration sensors: A numerical study

This study presents a novel approach to detect damage in steel frames using a hybrid network of piezoelectric strain and acceleration sensors. A numerical study has been conducted on a steel frame with bolted connections to verify the accuracy of the proposed method. The damage is introduced to the frame structure by loosening the bolts and creating cracks on its structural members. The frame is subjected to cyclic loading. Circular Lead Zirconate Titanate (PZT) piezoelectric transducers and bimorph PZT cantilever plates are used as strain and acceleration sensors, respectively. The strain and acceleration time histories are obtained from the finite element (FE) model. A theoretical model is used to obtain the voltage output delivered by the PZTs. Initial damage indicator features are defined by fitting a Gaussian mixture model (GMM) to the sensors output histograms. Moreover, a new sensor fusion model is proposed to improve the accuracy of the damage detection approach. The numerical results indicate that strain-based sensors and accelerometers are, respectively, more sensitive to cracks and bolt loosening. The hybrid system of sensors is efficient in detecting and localizing both types of damages in steel frames.