Commit to a grid-expansion plan you can prove is the best
A network-expansion plan proven optimal over every one of more than a million alternatives, plus the network’s real-time response simulated on quantum hardware.
Impact
The problem
Reinforcement plans are usually chosen by heuristics that return a good plan but cannot show it is the best one. That leaves the planner unable to defend the choice line by line.
Our approach
For a network-expansion case with twenty candidate builds we searched the whole space of plans and showed the chosen plan is the global optimum. Separately, we simulated the network’s real-time response on a 156-qubit processor, to 8 time steps where the classical simulator we ran stopped at 6.
The evidence
- Exhaustive search over 2^20 build plans; one basin, one global optimum (classical computation).
- Network response simulated to 8 time steps on a 156-qubit processor.
- Not claimed: any speed advantage for quantum in finding the plan.
Status: Simulation + Quantum hardware. Optimality proven by classical computation; the network response is measured on quantum hardware.