An under sampled impact location method based on FBG sensor

As one of the most critical tasks in structural damage monitoring, real-time impact localization plays the vital role in improving the durability of a structure, especially in the field of aerospace. Fiber Bragg Grating (FBG) sensors have been widely applied in composite materials structural health monitoring (SHM) system. This paper proposes a progressive combination recognition algorithm named back propagation-dictionary sparse representation-based classifier (BP-DSRC) to process signals obtained by FBG sensors in Carbon Fiber Reinforced Polymer composite SHM system, and to accomplish impact localization with higher accuracy. Considering the limited training set in the actual monitoring system, we specifically combine the back propagation (BP) neural network model with which a smaller range of positioning can be divided initially, sparse representation-based classifier (SRC) and K-means singular value decomposition (K-SVD) algorithm. Due to the role of SRC, the positioning effect can be more accurate than the signal matching algorithms. Training dictionary by K-SVD algorithm improves the positioning accuracy of SRC effectively. Meanwhile, considering the relatively low sampling rate of FBG sensors, the average and energy of signal are chosen as the input features of our impact localization algorithm. We implement the algorithm to an actual impact localization monitoring system with composite plate which shows that the proposed localization technique presented is an effective means of estimating impact locations.