Physics informed neural networks for shunted piezoelectric systems
Please login to view abstract download link
Shunted piezoelectric systems are used for vibration suppression in modern smart structures [1]. Optimal design of the system and fine tuning of parameters is a challenging task, including the determination of various uncertain parameters. Usage of machine learning has provided with useful tools in order to facilitate this task [2]. Physics informed neural networks will be investigated in this contribution for the solution of direct and inverse problems related to model piezoelectric systems, by following parallel developments in multi-layered piezoelectric micro-benders [3] and physics informed neural networks in elasticity [4]. Acknowledgements. Work supported by Hubert Curien 2023 bilatera scientific cooperation project. REFERENCES [1] K. Marakakis, G.K. Tairidis, P. Koutsianitis, and G.E. Stavroulakis, Shunt piezoelectric systems for noise and vibration control: a review. Frontiers in Built Environment, 5, 64, 2019. [2] G.A. Drosopoulos, G. Foutsitzi, M.-S. Daraki, G.E. Stavroulakis, G.E, Vibration Suppression of Graphene Reinforced Laminates Using Shunted Piezoelectric Systems and Machine Learning. Signals, 5, 326-342, 2024. https://doi.org/10.3390/signals5020017 [3] H. Binh H. Huy Nguyen et al, Physics-informed neural network with data-driven in modeling and characterizing piezoelectric micro-bender, J. Micromech. Microeng. in press https://doi.org/10.1088/1361-6439/ad809b, 2024. [4] A.D. Mouratidou, G.A. Drosopoulos, and G.E. Stavroulakis. Ensemble of physics-informed neural networks for solving plane elasticity problems with examples, Acta Mechanica, 1-20, 2024.
