Abstract:
A Structural Health Monitoring (SHM) system can continuously report on the structure’s status,
perform data analysis, and deliver assessment results automatically. SHM systems can play a very
important role, especially in aerospace structures in which a variety of low-weight materials i.e.
when composites are in use, where it is very necessary to know the health and degradation level
of such structures. To measure the health of such structures using SHM systems various
sensors/transducers are required. Piezoelectric sensors have been is used in SHM systems for the
past few decades because of their many advantages such as robustness, fast response time, compact
size, and versatility. A few active and passive technologies have been developed to make use of
piezoelectric sensors for example Guided waves (GW), Acoustic emission and Electromechanical
Impedance (EMI). The piezoelectric sensors have a vast variety of advantages but they are also
prone to some disadvantages during their usage. These disadvantages mostly occur due to certain
environmental effects like temperature, radiation, and vibration etc. Environmental effects and
damages to the structure play a significant role on the data acquisition by the sensors. In this paper,
we are trying to measure the effect of structural damage and temperature on the sensors used to
collect the data for an EMI-based SHM system and develop a technique using various Machine
Learning Algorithms to compensate for these effects.
