This contribution focuses on a detailed analysis of piezoelectric wafer active sensors, aiming to develop and evaluate inspection methodologies. Piezoelectric wafer active sensors are commonly used for active structural health monitoring systems. With the piezoelectric effect they can generate and sense waves, which can interact with the structure and with possible defects in the structure. The method of using these effects is known as acousto-ultrasonics. To enable a self-check of an active structural health monitoring system, an inspection of the piezoelectric wafer active sensors is necessary. Their different possible defects could cause a malfunction of a structural health monitoring system. Either damage of the structure is not detected or an undamaged structure is classified as damaged and false alarm is reported by the structural health monitoring system.
An analysis of the effects of defect piezoelectric wafer active sensors on the generated wave field is presented. It shows that the defects change the wave field in a non-negligible way. This change depends on defect type, defect severity and type of transducer. It has a significant influence on the results of the structural health monitoring system.
In this contribution the electro-mechanical impedance is carefully selected as a physical quantity, which enables to draw conclusions on possible sensor faults. A physical model is developed to show the impact of geometric and material parameters on the electro-mechanical impedance. A numerical model is constructed to calculate the electro-mechanical impedance and the generated wave field with the same model. Moreover, different types of defects are simulated. Experimental studies are set up to validate the models and quantify realistic parameter deviations for different application scenarios. Based on this knowledge, different new model-based and model-free methods for inspecting piezoelectric wafer active transducers are developed.
A novel approach to evaluate the merit of these methods is another focal point of this contribution. Based on statistical methods the performance is measured by combining the output of an inspection method with the effect of a defect on the result of the structural health monitoring system. When the application of this approach is carried out with numerical methods, the concept of model-assisted probability of detection is used. With the numerical model the electro-mechanical impedance as well as the generated wave field are simulated. The approach allows an application-based statement about the inspection method, used for a self-check.