Computing for Fuel-Cycle Optimization
The fuel cycle ties breeding, extraction, storage, and injection into one loop that must balance continuously; computing keeps it balanced.
The fuel cycle as a loop
A D-T plant's fuel cycle is a closed loop: tritium is bred, extracted, purified, stored, injected, partly burned, and the rest recovered. Every stage has losses and delays, and the loop must stay in balance so the plant never runs short of fuel or accumulates more than is safe.
What must balance
- Breeding rate against burn rate and losses.
- Extraction and purification throughput against demand.
- Stored inventory against safety limits.
- Startup inventory bred for later units.
Why it is a computing problem
The stages are coupled and time-dependent, with decay and permeation acting throughout, so keeping the loop balanced is a dynamic optimization. It draws on breeding models, inventory accounting, and supply-chain simulation, integrated into one model of the whole cycle.
The self-sufficiency requirement
The Hyperion breeder targets a breeding ratio of 1.8 precisely so the fuel cycle can sustain itself and build startup inventory for the next units. Optimizing the cycle turns that ratio into a workable, scheduled fuel supply rather than a static number.
Coupled to operation
The cycle responds to how the plant runs, so it is linked to operating scenarios and to the digital twin that tracks the real inventory in operation.
Honesty and safety
Fuel-cycle predictions carry the uncertainty of their loss and efficiency terms, carried through explicitly, and the inventory limits feed directly into safety and licensing.