The Global Centre for Risk and Innovation (GCRI)
A power grid is a single machine spread across a region or a continent: generation and demand must balance every second, and a fault anywhere is felt everywhere that is synchronously connected. Interconnection shares reserves and smooths variability, but it also shares disturbances. This page sets out how grids are operated within security limits, how failures propagate along lines and across borders, and what evidence operators and regulators use to plan and run them.
GCRI · Established · service level R0
Power grid reliability is the ability of a grid to deliver electricity within accepted standards and in the amount desired. Generation and demand must balance every second, and a fault anywhere is felt across the synchronous area. Operators run grids within security limits such as the N-1 criterion; interconnection shares reserves and also disturbances.
Where demand concentrates in cities and the rail, road and port corridors that move fuel and equipment, shown for Viet Nam as an example geography for this subject.
Transmission, distribution and system dependencies. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.
Frequency falls when demand exceeds supply and rises when it does not. Inertia from rotating machines and fast reserves arrest the change; as inverter-based generation grows, operators must procure these services explicitly rather than receive them for free.
Operators run the system so that the loss of any single major element, the N-1 criterion, leaves flows and voltages within limits. Operating close to those limits leaves no margin for a second event before the first is cleared.
An overloaded line trips, its flow shifts to parallel paths, those overload in turn, and protection separates the grid into islands with too much or too little generation. Cascades unfold in minutes and cross borders faster than operators can coordinate.
Control centres need telecommunications, substations need batteries, and gas plants need pipelines whose compressors may need electricity. Water, payments and hospitals depend on the grid in turn, so a grid failure becomes a multi-sector event.
Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.
| Term | Definition |
|---|---|
| N-1 criterion | The planning and operating rule that a grid must stay within limits after the loss of any single element, such as one line or generator. |
| SAIDI | The system average interruption duration index: the average outage minutes per customer over a year, defined in IEEE 1366. |
| SAIFI | The system average interruption frequency index: the average number of interruptions per customer over a year, defined in IEEE 1366. |
| Synchronous inertia | The kinetic energy stored in spinning generators that slows frequency change after a disturbance. |
Each indicator carries its unit, its cadence, the kind of source behind it and the evidence label a value would carry. No values are shown: a programme supplies them under its own authority.
| Indicator | Unit | Cadence | Source type | Label |
|---|---|---|---|---|
| System average interruption duration index (SAIDI) | minutes per customer per year | annual | Utility interruption records under IEEE 1366 | OFFICIAL |
| System average interruption frequency index (SAIFI) | interruptions per customer per year | annual | Utility interruption records under IEEE 1366 | OFFICIAL |
| Frequency deviation and time outside the standard range | hertz; minutes per month | continuous | Operator measurements, phasor measurement units | OBSERVED |
| Cross-border transfer capacity | megawatts | hourly | Transparency platform publications | OFFICIAL |
| Transmission circuit availability | percent of hours available | annual | Operator outage records | REPORTED |
| Congestion management and redispatch volume | megawatt hours; cost in national currency | monthly | Operator and regulator reports | OFFICIAL |
| Synchronous inertia | gigawatt seconds | hourly | Operator estimation from committed units | ESTIMATED |
The topic as a computational graph: named sources feed the methods, an evidence review gates the outputs, and the decisions stay with the authorities that hold them.
Fig. 02Nodes carry their type, two metric rows and a state; connectors show what feeds what; dashed frames name the owners; the timeline is the cadence. Design of the pipeline, not a record of a run.
Named public datasets and institutional records, each with its licence class. Sources by agreement are used only under their terms and never republished.
| Source | What it gives | Licence class |
|---|---|---|
| ENTSO-E Transparency Platform | Load, generation, cross-border flows, transfer capacities and outages for European bidding zones | Free, registration, ENTSO-E terms |
| EIA Hourly Electric Grid Monitor (Form EIA-930) | Hourly demand, generation and interchange for US balancing authorities | Public domain |
| EIA-861 annual electric power industry report | Utility reliability indices and customer counts | Public domain |
| ENTSO-E Ten-Year Network Development Plan | Planned transmission projects and system needs studies | Public report, ENTSO-E terms |
| OpenStreetMap power infrastructure | Lines, substations and plants mapped by contributors | Open Database Licence |
| Gridfinder | Modelled medium-voltage network extent from night-time lights and roads | Free and open, attribution |
| Incident investigation reports (ENTSO-E expert panels, NERC and FERC) | Reconstructions of major disturbances, their causes and recommendations | Public reports |
Energy Nexus prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.
| Who decides | The question | Evidence this page serves |
|---|---|---|
| Transmission system operators | Which contingencies to plan for and how much reserve and inertia to procure | Contingency scenarios, inertia estimates, lessons from published incident reports |
| Energy regulators | Whether to approve transmission investment and how to set reliability incentives | Reliability indices by region, congestion costs, network development plans |
| Regional coordination centres and neighbouring operators | How to share reserves and align defence plans across borders | Cross-border flow and capacity records; cascade scenarios that cross the border |
| Civil protection and critical service operators | Which services lose power first in a regional outage and what backup they need | Grid dependency briefs with backup endurance by site |
Electricity follows the physics of the network, not contracts. When a line trips, its flow redistributes instantly across parallel paths. If those paths overload, protection trips them too, and a sequence of trips can separate the grid into islands with mismatched supply and demand. Under-frequency load shedding then disconnects customers in blocks to keep the remaining system alive.
Inertia from spinning generators slows the rate at which frequency falls after a disturbance, giving controls time to act. Inverter-based resources such as wind, solar and batteries do not provide inertia inherently, although grid-forming controls can provide similar support. Voltage stability depends on reactive power close to where it is needed. Weak links between regions can produce oscillations that grow if they are not damped.
Steady-state models check line loading and voltages under normal conditions and after outages. They rely on accurate network data, which is often confidential.
Time-domain models study stability over seconds. Their accuracy depends on models of inverter controls that manufacturers may not fully disclose.
Network science studies the structure of grids and how failures spread. Topological models are useful for insight but ignore much of the physics.
Post-event reviews by operators and regulators remain the richest evidence, but they are published unevenly across jurisdictions.
The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.