Abstract Details
Abstracts
Author: Edward A Tocco
Requested Type: Poster
Submitted: 2026-03-16 14:18:06
Co-authors: R. Conlin, I. G. Abel
Contact Info:
University of Maryland
4298 Campus Drive
College Park, MD 20742
United States
Abstract Text:
With increasing focus on designing the core of a fusion pilot plant, the focus of simulations has moved from highly accurate verification and validation simulations to the task of trying to design and optimize whole devices. In particular, we wish to examine a small number of global figures-of-merit that characterize the performance of such a device, such as stored energy or fusion yield. Previous efforts [e.g. P. Kim et. al. J. Plas. Phys. 2024] have focused on optimizing a parameter (e.g. the turbulent heat flux) at a single point produced some improvement in profile characteristics. However, the interrelated and nonlinear nature of the system means that optimizing the parameters at one point is unreliable as a method for improving overall performance. This motivates efforts to instead optimize these global figures-of-merit using a global simulation. In this aim, we use the multiscale plasma transport framework [Abel et. al. 2013 Rep. Prog. Phys.]. Because these computation of fluxes for transport simulations often involves expensive gyrokinetic simulations, we wish to use efficient gradient-based optimization schemes to minimize the number of calls to external gyrokinetic codes. Adjoint sensitivity analysis is well suited to this task, as it allows for calculation of the gradients of the figure of merit with respect to an arbitrary of parameters with minimal additional cost [Plessix et. al. Geophys. J. Intl. 2006].
MaNTA [Abel et. al. in progress], a new code designed to solve an arbitrary system of 1D nonlinear reaction-diffusion equations, was modified to compute adjoint sensitivities. We use MaNTA coupled with the neoclassical code YANCC [Conlin in progress] and equilibrium code DESC [Dudt et. al. Phys. Plas. 2020] to maximize the stored energy of the plasma. We compare these results with other methods of stellarator optimization, e.g. quasisymmetry.
Characterization: 4.0
Comments: