CAREER: Guided Exploration of Multiphysics Design Space for Electric Machines Using Tensorial Analysis (GEOMETRY)

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

The year 2023 has witnessed the hottest days on Earth since 1940, highlighting the urgency of the fight against climate change and excessive carbon emissions. In this mission, motors and generators, or collectively electric machines (EMs), play a pivotal role. Through EMs, over 90% of global electricity is generated, 45% of which is turned into mechanical work. EMs are at the heart of electric vehicles (EVs), wind energy generation, and various industrial processes, propelling these essential applications forward. To push EM performance boundaries, it is crucial to explore the multiphysics design space, which encompasses the interplay of various physical disciplines, such as electrodynamics, heat transfer, and structural mechanics. The multiphysics design space is confined by EM topologies, i.e., arrangements of constituent parts in EMs including steel, copper, magnets, etc. A natural question to ask is: what are the best arrangements of these constituent parts? The proposed research is poised to systematically answer this fundamental question and accelerate the exploration of high-performance and highly sustainable EMs. Parallel to the research, the PI’s education goal is to systematically foster diverse multiphysics designers, including those who are underrepresented. The PI’s education activities will include constructing a website-based learning platform named TENSOR to create an inclusive multiphysics learning hub, implementing a duality and analogy-based multiphysics education technique for K-12, undergraduate, and graduate students, and offering an open-access new EM design course to the frontier EM workforce. In the existing design paradigm, new EM topologies are often conceived by designers and the conception relies on their intuition, resulting in sporadically revealing new design space and corresponding performance space. To overcome the limitations of this intuition-based design paradigm, the overarching goal of this CAREER project is to pioneer a new design paradigm — multiphysics synthesis which realizes guided exploration of multiphysics design space for EMs using tensorial analysis. To achieve this goal, three research thrusts are planned. First, primitive EMs, governed by electrodynamics and expressed in hyperdimensional space, will be used as starting points to derive new topologies. Second, the hyperdimensional models of EMs will be embodied in the 3D space for performance evaluation and optimization. Third, the methodologies in the first two thrusts will be generalized to incorporate physics beyond electrodynamics and fulfill multiphysics synthesis. In all three thrusts, tensorial analysis originated from mathematics (geometry) and theoretical physics (relativity theory) will be applied. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.