liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Large Language Model in Aircraft System Design
Linköping University, Department of Management and Engineering, Fluid and Mechatronic Systems. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0002-2315-0680
2024 (English)In: ICAS PROCEEDINGS 34th Congress of the International Council of the Aeronautical Sciences Florence, Italy / [ed] Chris Atkin, 2024Conference paper, Published paper (Other academic)
Abstract [en]

The introduction of large language models in the form of ChatGPT at the end of 2022, marked the beginningof the widespread democratization of artificial intelligence. This is spawning a entirely new area of researchinto application of this technology. This paper presents a novel approach to generate configuration rules, anddesign concepts using ChatGPT. The objective is to demonstrate how large language models models can aidin automating the engineering design process. In this paper the configuration of aircraft hybrid propulsionsystems is studied as an example. It is shown how a system configuration can be generated and presentedin the form of UML component diagrams. Large Language Models usually have an element of randomnessinserted in their response. This makes them inherently non-deterministic, and the result is not always correct.Therefore, statistical properties are studied such that the probability of getting a correct result can be estimated,and prompts can be tweaked to provide the best result. This is particularly useful when a prompt is separatedinto a general part that can be reused for similar tasks, and one specific part.It is also shown how these models can used in consecutive step to formulate a simulation model and executeit within the framework of ChatGPT-4, or be exported to more dedicated simulation models for more advancedsimulations. Also optimization can be included. The study shows promising results, and it shows that thereis an immense potential for AI in engineering system design, and we are just in the beginning of applying thistools.

Place, publisher, year, edition, pages
2024.
Keywords [en]
Large language models, aircraft design, aircraft systems, propulsion systems, modelling and simulation
National Category
Vehicle and Aerospace Engineering
Identifiers
URN: urn:nbn:se:liu:diva-211541OAI: oai:DiVA.org:liu-211541DiVA, id: diva2:1935885
Conference
34th Congress of the International Council of the Aeronautical Sciences Florence, Italy
Available from: 2025-02-08 Created: 2025-02-08 Last updated: 2025-02-14

Open Access in DiVA

No full text in DiVA

Authority records

Krus, Petter

Search in DiVA

By author/editor
Krus, Petter
By organisation
Fluid and Mechatronic SystemsFaculty of Science & Engineering
Vehicle and Aerospace Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 32 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf