Abstract
Edge plasma turbulence is critical to the performance of magnetic confinementfusion devices. Towards better understanding edge turbulence in both theory andexperiment, a custom-built physics-informed deep learning framework constrainedby partial differential equations is developed to accurately learn turbulentfields consistent with the two-fluid theory from partial observations ofelectron pressure. This calculation is not otherwise possible usingconventional equilibrium models. With this technique, the first directquantitative comparisons of turbulent fields between electrostatic two-fluidtheory and electromagnetic gyrokinetic modelling are demonstrated with goodoverall agreement found in magnetized helical plasmas at low normalizedpressure. To translate these computational techniques to experimental fusion plasmas, anovel method to translate brightness measurements of HeI line radiation intolocal plasma fluctuations is demonstrated via a newly created deep learningframework that integrates neutral transport physics and collisional radiativetheory for the $3^3 D - 2^3 P$ transition in atomic helium. Using fast cameradata on the Alcator C-Mod tokamak, this thesis presents the first 2-dimensionaltime-dependent experimental measurements of the turbulent electron density,electron temperature, and neutral density in a fusion plasma using a singlespectral line. With this experimentally inferred data, initial estimates of the2-dimensional turbulent electric field consistent with drift-reduced Braginskiitheory under the framework of an axisymmetric fusion plasma with purelytoroidal field are calculated. The inclusion of atomic helium effects onparticle and energy sources are found to strengthen correlations between theelectric field and electron pressure while broadening turbulent fieldamplitudes which impact ${\bf E \times B}$ flows and shearing rates.