How it works
Availability can be measured by time or by energy. Time-based availability counts the hours a unit could run and divides by the hours in the period. Energy-based availability compares the energy the plant produced with the energy it could have produced had every unit been running. For a wind farm, that means estimating, for each turbine and each ten-minute period, what it would have made from the measured wind and its power curve.
The rules on what counts matter as much as the formula. Contracts list exclusions such as grid outages, curtailment by the grid operator, force majeure, scheduled maintenance within an agreed allowance, and sometimes low-wind or night-time hours. Each period is classified from SCADA status codes and alarm logs. A turbine stopped by its own fault counts against availability. One stopped by a grid instruction does not. Disputes between owner and contractor usually come down to this classification.
Why it matters
Availability is what O&M contractors and turbine and inverter suppliers guarantee, typically around 97% to 99%, with payments for shortfalls. It separates losses the contractor controls from losses caused by the weather or the grid. For an owner, each point of energy-based availability is close to a point of revenue. For a lender, it is a key assumption behind the P90 energy estimate.
A worked example
A 30-turbine wind farm runs through a 720-hour month, giving 30 × 720 = 21,600 turbine-hours. Faults stop turbines for 380 hours. Planned maintenance takes 120 hours, of which the contract excludes 100. Grid outages cause 200 hours, all excluded. Excluded time is 100 + 200 = 300 hours, and counted downtime is 380 + 20 = 400 hours. Availability is (21,600 − 300 − 400) ÷ (21,600 − 300) = 20,900 ÷ 21,300 = 98.1%.
Illustrative figures.
Across markets
In India, availability is also a tariff term. Storage and round-the-clock tenders pay or penalise against a contracted availability, for example 95% for batteries. In the US, capacity markets such as PJM's charge penalties when a committed resource is unavailable during system stress. In Europe, availability mostly lives in supplier and O&M contracts, and in the performance rules of reserve products.