Electro-mechanical impedance based damage detection has been shown to detect small changes in structure. Due to the large frequency band which may be assessed with the EMI technique, it has been shown to detect very local damage irrespective of the change in the boundary conditions. Studies have also been carried out to compensate for ambient temperature effects. Unfortunately, the drawback of the local nature of the EMI approach is the relatively low range of sensing. As a result most of the studies using EMI approach are limited to applications where the sensitive region is apriori known. This is not always the case, and thus the study of the technique at the array level is necessary. Thus, the present study tries to establish the approach for optimization of the array of PZTs for EMI based approach. The primary task is to establish the range of the sensors to damage in an anisotropic GFRP plate. The range of the sensors and their directionality then may be used to optimize the sensor placement to ensure the maximum coverage of the plate for damage detection. Once the cost function of the optimization has been established, genetic algorithm (GA) is employed for optimization. GA offers several advantages over brute force based methods as well as other optimization approaches. GA is ideally suited for multi-objective optimization which then paves the way for incorporating other optimization objectives in the search.
|