Underperformance
A turbine producing less than it should in the conditions it actually experienced, as distinct from producing less than a neighbor or less than last year.
Underperformance is a comparison, and the comparison it is made against decides whether the word means anything. Against last month it mostly measures the weather. Against the site average it penalizes machines in poor positions and excuses machines in good ones. Against a datasheet power curve it measures the age and siting of the turbine as much as its condition.
The version that is actionable compares a machine against what it should have produced given the wind speed, direction, air density, and turbulence intensity it actually saw. That expectation is built from the turbine's own history, so the residual - the gap between expected and measured - is a property of the machine rather than of the site.
Persistent underperformance in the partial-load region usually has a small number of explanations: a control offset such as yaw or pitch, a degraded blade surface, a sensor reading wrong, or a component beginning to fail. Each carries a different cost and a different intervention, and separating them is the work of loss attribution.
The reason it is worth chasing is arithmetic. A turbine losing two percent of its output raises no alarm and appears in no report, but two percent across a fleet, sustained for a year, is a large number of megawatt-hours - and unlike new capacity, recovering it requires no capital project.
A more intelligent approach to sustainable energy.
See how OpenTurbine can help your team understand performance, anticipate operational issues, and make better decisions across your wind assets.