Evaluation of plates in similitude by experimental and machine learning techniques

The dynamic behaviour of structures can be investigated by using the concepts of complete and partial similitudes, being the latter much more of interest since the complete similitudes are difficult to be achieved and the experiments are often executed by using distorted models as test articles. In this work, beams and plates in similitude have been investigated by using Machine Learning (ML) to predict the dynamic response characteristics without invoking governing equations and/or solution schemes. ML is based on algorithms that derive models from sample inputs providing data-driven prediction. The absence of an explicit algorithm, being the process totally data-driven, confers to the approach a high versatility which allows its application even in the vibroacoustic research fields and problems. In view to validate the ML predictions, numerical investigations of beams and plates in similitude have been performed. The good predictions obtained with ML highlights the potentialities of these algorithms and open the way to analyses with more complex structures.