Tien Vuong Nguyen

In the era of data and artificial intelligence (AI), data-driven models are revolutionizing the way we simulate and analyze dynamic systems. Deep Neural Networks (DNNs), a powerful subset of AI, are increasingly being employed to model complex mechanical systems, offering a transformative alternative to traditional physics-based approaches. A number of mathematical models have been developed to represent the physics brought on by factors such as friction, clearance, and nonlinear interactions, but these models are often highly complex and require accurate identification of each variable in the equations, which can be challenging. By leveraging large datasets, DNNs can learn the underlying patterns and relationships within dynamic systems, enabling accurate predictions and simulations without the need for explicit mathematical formulations. In mechanical engineering, where systems often involve nonlinearities, friction, and complex interactions, data-driven models provide a flexible and efficient solution. DNNs can capture intricate dynamics, such as vibrations, wear, and energy dissipation, by training on experimental or simulation data. This capability is particularly valuable in applications like robotics, vehicle dynamics, and machinery, where real-world performance depends on understanding system behavior under varying conditions.