<title>Automated real-time structure health monitoring via signature pattern recognition</title>

TLDR

The study addresses critical challenges of sensing area localization and sensor temperature drift in structural health monitoring, presenting a novel temperature compensation method based on probability correlation functions. The paper details an automated real‑time structural health monitoring system. The system employs piezoceramic patch arrays as integrated sensor‑actuators, an impedance analyzer for frequency‑response acquisition, and a PC for control and display, using structural signature pattern recognition on a 3‑bay truss test bed while incorporating temperature compensation via probability correlation functions. Because of its lightweight, compact sensor array and incipient‑damage sensing capability, the system is suitable for real‑time health monitoring of aircraft, spacecraft, large‑span roofs, and steel bridges.

Abstract

Described in this paper are the details of an automated real-time structure health monitoring system. The system is based on structural signature pattern recognition. It uses an array of piezoceramic patches bonded to the structure as integrated sensor-actuators, an electric impedance analyzer for structural frequency response function acquisition and a PC for control and graphic display. An assembled 3-bay truss structure is employed as a test bed. Two issues, the localization of sensing area and the sensor temperature drift, which are critical for the success of this technique are addressed and a novel approach of providing temperature compensation using probability correlation function is presented. Due to the negligible weight and size of the solid-state sensor array and its ability to sense incipient-type damage, the system can eventually be implemented on many types of structures such as aircraft, spacecraft, large-span dome roof and steel bridges requiring multilocation and real-time health monitoring.