B. Zorawski, S. Burela, P. Krah, A. Marmin, K. Schneider
ECCOMAS 2024.
Abstract
This paper presents a neural network-based methodology for the decomposition of transport-dominated fields using the shifted proper orthogonal decomposition (sPOD). Classical sPOD methods typically require an a priori knowledge of the transport operators to determine the co-moving fields. However, in many real-life problems, such knowledge is difficult or even impossible to obtain, limiting the applicability and benefits of the sPOD. To address this issue, our approach estimates both the transport and co-moving fields simultaneously using neural networks. This is achieved by training two sub-networks dedicated to learning the transports and the co-moving fields, respectively. Applications to synthetic data and a wildland fire model illustrate the capabilities and efficiency of this neural sPOD approach, demonstrating its ability to separate the different fields effectively
Abstract This paper presents a neural network-based methodology for the decomposition of transport-dominated fields using the shifted proper orthogonal decomposition (sPOD). Classical [...]
N. Shahmansouri, H. Cheong, A. Tessier, A. Butscher
ECCOMAS 2024.
Abstract
In kinematic mechanism synthesis, the goal is to find the optimal configuration and parameters of a mechanism system that produces desired mechanical performance such as motion or force. For a problem involving a complex set of requirements, the optimal system often comprises of many mechanism components, known as Mechanical Building Blocks (MBBs). For example, a complex power transmission system is created with a series of gears, shafts, belts, etc. During the search for an optimal system, the algorithm must be able to evaluate the perfor mance of a candidate system made up of an arbitrary collection of building blocks. To address this challenge, we propose modular modelling of the MBBs that can be composed on-the-fly as a system of equations to be solved. This approach is largely based on multidisciplinary design optimization framework, where the model is composed by considering all relevant disciplines simultaneously to find an optimal solution. In this work, we present the first set of MBBs modelled so far, and three use cases where these building blocks are automatically composed to create a complex mechanism system and analyzed to find the optimal parameters of the system. Our approach is implemented using Dymos, which employs modular analysis and unified derivatives (MAUD) for computing the total derivatives out of the partial derivatives of individual building blocks for gradient-based optimization and a direct collocation method for integrating the kinematic equations. In sum mary, our work demonstrates the value of the multidisciplinary design optimization approach in solving a mechanism synthesis problem
Abstract In kinematic mechanism synthesis, the goal is to find the optimal configuration and parameters of a mechanism system that produces desired mechanical performance such as motion [...]