Tese: Nonlinear system identification of hybrid machine learning and physical models for mechanical systems
Aluno(a) : Daniel Henrique Braz de SousaOrientador(a): Helon Ayala
Área de Concentração: Mecânica Aplicada
Data: 06/03/2023
Link para tese/dissertação: https://doi.org/10.17771/PUCRio.acad.62550
Resumo: There is a growing demand for accurate dynamic models, driven by the Industry 4.0 paradigm that introduces, among others, the concept of the digital twin in which dynamic models play an important role. Ideally, a dynamic model presents a compromise between complexity and accuracy, while providing physical interpretability about the system. Aiming at these characteristics, this work proposes a hybrid identification methodology that combines a gray-box phenomenological model with a black-box model based on artificial neural networks. The proposed methodology is applied in three case studies of nonlinear systems with experimental data, namely, the vertical dynamics of a vehicle, an elastomer-based series elastic actuator, and an electromechanical positioning system. The results show that the proposed hybrid model is up to 60\% more accurate while providing the physical interpretability of the system, without significantly increasing the complexity of the model.