AI-Driven Design-Space Exploration for Thermo-Fluid Domains

Amit Bhatia, Kathryn Kirsch, Claudio Pinello, Ram Ranjan

Keywords: Design Space Exploration, Constraint Satisfaction Problem, Artificial Intelligence, Aerospace, Machine Learning

TL;DR: In this paper, we present an AI-driven design space exploration framework for thermo-fluid domains.

Abstract: Existing approaches to exploring design space in thermo-fluid domains are human-labor intensive and not easily automatable. As a result they can typically generate only minor variations of pre-existing designs. In this paper, we discuss results from an AI-assisted design-space exploration tool built at our organization for such domains.