Strain sensor's network for low-velocity impact location estimation on carbon reinforced fiber plastic structures: Part-II.

Identification of low velocity impact (LVI) location in composite aircraft structures is seamless need for safe, reliable operation and maintenance of aerospace industry. To locate the LVI’s an optimized sensor network has designed using the strain response from fiber Bragg grating (FBG) & resistance strain gauge (RSG) sensor bonded to the composite structure. Strain scan (SS) algorithm has been developed to locate such events reported as Part-I. In this work, we have developed a novel algorithm based on weighted energy (WE) of the sensor response. The LVI’s has been carried out on composite structures & the locations of LVI’s have estimated using SS, WE & previously developed machine learning base support vector machine (SVM) algorithms. The WE and SS algorithms are based on proximity of events (closer to the sensor, higher the response), whereas LS-SVR is a data-driven approach. Further, we have compared the performance of the developed algorithms and algorithms cited in the literature using the performance index (PI), a measure of estimation efficiency as a function of the number of sensors, dimension/area of the structure, error & number of test cases. It is established that WE algorithm shown suprema performance over the other algorithm with 34 mm mean Euclidian distance error & PI value of 5.5