The day's work, in order
Findings arrive ranked by the energy and revenue at stake, so the first task of the morning is not deciding what the first task is. The effect is largest on the work that would otherwise never reach the top of a list at all.
A ranked answer to the question of which turbine to look at today.
An operating team has more machines than hours. What sets annual energy production over a year is not the response to the failures, which is generally good, but the order in which everything else receives attention.
What an operating team receives each day
Not another dashboard to interpret. A shorter list of work, in the order that returns the most.
Corrections that need no site visit
Static yaw misalignment and pitch calibration drift are corrected in the controller. They raise no fault, they persist for years once introduced, and they are among the lowest-cost production gains available on an operating site.
Weather separated from machine
A week of low output is either a low wind week or a real loss, and the distinction cannot be made by looking at production. Comparing each turbine against machines in the same conditions is what makes it.
Less time spent proving a problem exists
Establishing whether a machine is genuinely underperforming is the work an engineer currently completes before diagnosis can begin. That step is what the models do.
The largest losses on an operating site never generate a ticket.
An operations team responds to what the fleet reports. Faults raise alarms, outages raise work orders, and both are handled competently on a well-run site. Nothing raises a ticket for a turbine that is running, available, and producing four percent below the machines on either side of it.
Those losses are individually small and collectively larger than the outages, because they run continuously and across many assets. A yaw offset left in place after a service visit produces a deficit every hour the machine turns, for years, and the only trace it leaves is in a comparison nobody is currently making.
Prioritization on most sites is driven by escalation rather than by value.
The machine that failed, the machine the owner asked about, the machine due for scheduled service: each of those is a reasonable trigger, and none of them is the energy at stake. The result is a queue that is defensible task by task and suboptimal across a season.
When every finding carries the annual energy and revenue behind it, the sequence becomes an operating decision that can be explained to an owner rather than a queue that has to be justified after the fact. It also makes deferral explicit: work that is deliberately not done this quarter has a number attached to that choice.
The corrections worth finding first are settings, not components.
Control-side losses are the most attractive class of finding on an operating site because nothing is broken. A static yaw misalignment of eight to ten degrees costs a low single-digit percentage of that machine's annual energy, raises no alarm, and is corrected in the controller rather than at height. A collective pitch offset behaves the same way and is commonly introduced by a blade replacement or a software update.
They persist because the turbine believes it is behaving correctly. The sensor reports the wind as the machine has been told to read it, the controller acts on that reading, and every internal check passes. The error is only visible from outside the machine, in a comparison against units experiencing the same weather.
Distinguishing them from a position effect is the part that requires care. A turbine standing behind three others in the prevailing direction produces less for reasons of wake and array geometry, not condition, and sending an engineer to correct a fault it does not have is how a program loses a team's confidence in its first month.
Nothing about the site has to change to adopt it.
The platform does not replace the SCADA system, the manufacturer's portal, or the maintenance management system. It reads the record those systems already produce and returns a ranked set of findings into the place the team schedules work.
Nothing is installed at the turbine and nothing changes in the control system, so adoption requires no outage, no commissioning window, and no conversation with the manufacturer about certification. A first assessment runs on a historical export and produces findings the team can check against a period whose outcome they already know.
What changes on site
The measure of whether this is working is not the size of the report. It is whether the first task of the morning changed.
- A daily work list ordered by the energy at stake
- Control-side corrections identified without an intervention at height
- Weather separated from machine before an engineer is dispatched
- Underperformance evidenced per turbine rather than per site
- No new hardware, no controller changes, and no outage to adopt
The vocabulary on this page
Every term below has a full definition in the wind energy glossary, written for the people accountable for a fleet rather than for a search engine.
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.