A power grid is a graph of substation nodes connected by lines. When a node fails, its load must be picked up by neighbours through redundant links, or the overload cascades further.
Uptime = nodes_up / nodes_total
P(cascade) = severity × sensitivity / (1 + redundancy)
- Redundancy links — extra cross-connections between substations that allow rerouting around a failed node.
- Failure severity — how damaging the triggering event is to the first nodes it strikes.
- Repair speed — how quickly crews restore a failed node back online.
- Cascade sensitivity — how likely an overloaded neighbour is to fail in turn, propagating the outage.
Click Storm or Cyberattack to trigger an initial failure at a random node/cluster, then watch red (down) nodes attempt to reroute power through green (up) neighbours via the redundancy links.