English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The interaction of neural networks with physical equations offers a widerange of applications. We provide a method which enables a neural network totransform objects subject to given physical constraints. Therefore an U-Netarchitecture is used to learn the underlying physical behaviour of fluid flows.The network is used to infer the solution of flow simulations, which will beshown for a wide range of generic channel flow simulations. Physical meaningfulquantities can be computed on the obtained solution, e.g. the total pressuredifference or the forces on the objects. A Spatial Transformer Network withthin-plate-splines is used for the interaction between the physical constraintsand the geometric representation of the objects. Thus, a transformation from aninitial to a target geometry is performed such that the object is fulfillingthe given constraints. This method is fully differentiable i.e., gradientinformations can be used for the transformation. This can be seen as an inversedesign process. The advantage of this method over many other proposed methodsis, that the physical constraints are based on the inferred flow fieldsolution. Thus, we have a transferable model which can be applied to varyingproblem setups and is not limited to a given set of geometry parameters orphysical quantities.

Transferable Model for Shape Optimization subject to Physical Constraints