Institutsseminar/2021-12-17: Unterschied zwischen den Versionen

Aus SDQ-Institutsseminar
(Die Seite wurde neu angelegt: „{{Termin |datum=2021-12-17T11:30:00.000Z |raum=https://kit-lecture.zoom.us/j/67744231815 }}“)
 
Keine Bearbeitungszusammenfassung
Zeile 1: Zeile 1:
{{Termin
{{Termin
|datum=2021-12-17T11:30:00.000Z
|datum=2021-12-17T11:30:00.000Z
|raum=https://kit-lecture.zoom.us/j/67744231815
|raum=Raum 348 (Gebäude 50.34), https://conf.dfn.de/webapp/conference/979160755
}}
}}
The presentation will be held in person in Room 348, and it will be streamed via the DFN Conf Tool at the same time.

Version vom 13. Dezember 2021, 11:25 Uhr

Termin (Alle Termine)
Datum Freitag, 17. Dezember 2021
Uhrzeit 11:30 – 11:50 Uhr (Dauer: 20 min)
Ort Raum 348 (Gebäude 50.34), https://conf.dfn.de/webapp/conference/979160755
Webkonferenz
Vorheriger Termin Fr 10. Dezember 2021
Nächster Termin Fr 14. Januar 2022

Termin in Kalender importieren: iCal (Download)

Vorträge

Vortragende(r) Jonas Zoll
Titel Injection Molding Simulation based on Graph Neural Networks
Vortragstyp Proposal
Betreuer(in) Moritz Renftle
Vortragsmodus
Kurzfassung Injection molding simulations are important tools for the development of new injection molds. Existing simulations mostly are numerical solvers based on the finite element method. These solvers are reliable and precise, but very computionally expensive even on simple part geometries. In this thesis, we aim to develop a faster injection molding simulation based on Graph Neural Networks (GNNs). Our approach learns a simulation as a composition of three functions: an encoder, a processor and a decoder. The encoder takes in a graph representation of a 3D geometry of a mold part and returns a numeric embedding of each node and edge in the graph. The processor updates the embeddings of each node multiple times based on its neighbors. The decoder then decodes the final embeddings of each node into physically meaningful variables, say, the fill time of the node. The envisioned GNN architecture has two interesting properties: (i) it is applicable to any kind of material, geometry and injection process parameters, and (ii) it works without a “time integrator”, i.e., it predicts the final result without intermediate steps. We plan to evaluate our architecture by its accuracy and runtime when predicting node properties. We further plan to interpret the learned GNNs from a physical perspective.
Neuen Vortrag erstellen

Hinweise

The presentation will be held in person in Room 348, and it will be streamed via the DFN Conf Tool at the same time.