An investigation into the temperature stability of a guided wave structural health monitoring system using permanently attached sensors

When testing complex structures using a deployable guided wave system, the measured signals are often far too complex to directly interpret without a priori knowledge of the geometry of the structure. However, in a structural health monitoring (SHM) scenario, the sensors are permanently attached to the structure and so changes in the system can be monitored by comparison to an earlier baseline measurement, taken when the structure was undamaged. This paper investigates such a system for the SHM of plate-like structures using guided waves. The basis of the system is a distributed array of permanently attached piezoelectric guided wave transducers where pairs of transducers are used in pitch-catch configuration allowing the detection and localization of damage in the plate. It has been observed that the use of a single baseline measurement suffers in the medium to long-term due to the variation of environmental conditions. A key parameter in the instability is the change in the temperature of the test structure. A method of using a database of baselines, termed optimal baseline subtraction (OBS), is described and applied to an experimental SHM system. The objective of this paper is to investigate the suitability and robustness of OBS under varying rates of temperature change and its use in the localization of damage. Results have shown that through the OBS an improvement of up to 20 dB in signal-to-coherent noise ratio may be achieved compared with single baseline subtraction