An experimental technique for structural diagnostic based on laser vibrometry and neural networks

In recent years damage detection techniques based on vibration data have be en largely investigated with promising results for many applications. In pa rticular, several attempts have been made to determine which kind of data s hould be extracted for damage monitoring. In this work Scanning Laser Doppler Vibrometry (SLDV) has been used to dete ct, localise and characterise defects in mechanical structures. After dedic ated post-processing, a neural network has been employed to classify LDV da ta with the aim of automating the detection procedure. In order to demonstrate the feasibility and applicability of the proposed t echnique, a simple case study (an aluminium plate) has been approached usin g both Finite Element simulations and experimental investigations. The prop osed methodology was then applied for the detection of damages on real case s, as composite material panels. In addition, the versatility of the approa ch was demonstrated by analysing a Byzantine icon, which can be considered as a singular kind of composite structure. The presented methodology has proved to be efficient to automatically recog nise defects and also to determine their depth in composite materials. Furt hermore, it is worth noting that the diagnostic procedure supplied correct results for the three investigated cases using the same neural network, whi ch was trained with the samples generated by the Finite Element model of th e aluminium plate. This represents an important result in order to simplify and shorten the procedure for the training set preparation. which often co nstitutes the main problem for the application of neural networks on real c ases or in industrial environments.