Method for removing temperature effect in impedance-based structural health monitoring systems using polynomial regression

Structural health monitoring systems are employed to evaluate the state of structures to detect damage, bringing economical and safety benefits. The electromechanical impedance technique is a promising damage detection tool since it evaluates structural integrity by only measuring the electrical impedance of piezoelectric transducers bonded to structures. However, in real-world applications, impedance-based damage detection systems exhibit strong temperature dependence; therefore, variations associated with temperature changes may be confused as damage. In this article, the temperature effect on the electrical impedance of piezoelectric ceramics attached to structures is analyzed. Besides, a new methodology to compensate for the temperature effect in the electromechanical impedance technique is proposed. The method is very general since it can be applied to nonlinear (polynomial) temperature and/or frequency dependences observed on the horizontal and vertical shifts of the impedance signatures. A computer algorithm that performs the compensation was developed, which can be easily incorporated into real-time damage detection systems. This compensation technique is applied successfully to two aluminum beams and one steel pipe, minimizing the effect of temperature variations on damage detection structural health monitoring systems in the temperature range from −40°C to 80°C and the frequency range from 10 to 90 kHz.