Anzeige
Erschienen in:
2022 | OriginalPaper | Buchkapitel
Reducing Fuel Consumption by Virtually Testing an Engine with AI
verfasst von : Joël Henry, Tilmann Oestreich
Erschienen in: 22. Internationales Stuttgarter Symposium
Verlag: Springer Fachmedien Wiesbaden
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Among the multiple applications of machine learning for engineering and product development, we have recently used the Monolith AI platform to perform what we call ‘virtual testing’. Test campaigns can be expensive, but also time consuming. And sometimes, after all tests are completed, one might realise that they would have liked to make more tests. That is where virtual testing can help. The concept is to train machine learning models on existing test data and use the trained models to predict the outcome of other—virtual—test campaigns.
This has multiple applications: i) one can virtually predict tests that could have been done and save the cost and the time of those tests, and ii) one can virtually predict tests that cannot be done at that time, to gain more insight on potential/future designs and help deciding on the best strategy.
These two applications were explored on test data for engine calibration provided by Kistler, and in both applications insights could be gained by the engineering team. For both applications, the key advantage here is that although engineers might guess the expected trend of a test (e.g., less friction yields lower fuel consumption), such virtual test campaigns will add a missing yet critical piece of information, which is a quantitative value. Now the engineer knows for instance that the product can consume 2 %, or 0.7 % less fuel, and they can quickly make appropriate decisions.