Artificial Neural Networks in Structural Mechanics
Key Info
Basic Information
- Degree:
- Master
- Semesters:
- Wintersemester
- Lecturer:
- Prof. Dr.-Ing. Marcus Stoffel
- Language:
- English
Further Information
The students will learn to work with artificial neural networks from the viewpoint of engineering science. This implies to understand different network topologies and their applications in structural mechanics. Classical structural models will be replaced by artificial neural networks partly or completely depending on the current problem.
Students will learn to program neural networks and to train them by data gained from experiments or simulations. The aim is to increase the efficiency of structural calculations towards faster simulations and new structural models without material parameters trained just by experimental or simulated data. At the end of the course students will be able to use artificial neural networks for numerical predictions in structural mechanics. This includes the combination of neural networks with the finite element method. In this way, intelligent elements are developed.
Content
- work with artificial neural networks from the viewpoint of engineering science
- understand different network topologies and their applications in structural mechanics
- Classical structural models will be replaced by artificial neural networks partly or completely depending on the current problem
- program neural networks and to train them by data gained from experiments or simulations
- increase the efficiency of structural calculations towards faster simulations and new structural models without material parameters trained just by experimental or simulated data
- use artificial neural networks for numerical predictions in structural mechanics
- combination of neural networks with the finite element method