Compressive sensing based data-loss recovery enables robust estimation of damage index in ultrasonic structural health monitoring

Abstract

This work presents a demonstration of orthogonal matching pursuit (OMP) algorithm to recover lost data packets in ultrasonic sensor based structural health monitoring (SHM) applications. OMP leverages the sparsity of ultrasonic sensor signals shaped by a Hanning window to recover information from lossy transmission. To illustrate a case study for disbond detection application in presence of data loss for SHM of honeycomb composite sandwich structures (HCSS), we emulate data loss fractions varying from 10%-90% on a dataset comprising of 8 ultrasonic transducers and use OMP to recover the signals from the lossy datasets thus generated. Damage index (DI) estimations performed using the recovered signals indicate that OMP is a robust and reliable signal recovery technique, producing estimation errors in DI calculation consistently <1.5%.