Master Algorithmic Plasma Physics: AI-Accelerated Nuclear Fusion Commercialization in 4 weeks through hands-on, project-based online training with DSTC.
Data Science & Analytics
Module-by-module breakdown of Algorithmic Plasma Physics: AI-Accelerated Nuclear Fusion Commercialization, from foundations to a certified capstone project.
Plasma
โข Magnetic confinement configurations: tokamak and stellarator geometry
โข MHD equilibrium, transport and the instabilities that limit performance
โข Diagnostics: what is actually measured inside a burning plasma, and how sparsely
Simulation
โข Gyrokinetic and MHD codes and where their expense becomes prohibitive
โข Surrogate modelling to replace inner loops of expensive simulation
โข Validation against experiment rather than against other simulations
Control
โข Shape and position control as a control problem with hard safety limits
โข Reinforcement learning trained in simulation and the sim-to-real gap
โข Disruption prediction and mitigation within actuator response times
Design
โข Stellarator coil optimisation and high-dimensional design spaces
โข Multi-objective trade-offs across confinement, engineering and cost
โข Uncertainty propagation from physics models into design decisions
Commercialisation
โข Tritium breeding, materials survivability and the neutron problem
โข Reading a fusion company's technical claims critically
โข Where machine learning genuinely shortens the path and where it does not
e-Certificate and e-Marksheet issued on successful completion.