Recovering the production a wind fleet was built to deliver.

OpenTurbine applies machine learning to the operational data wind farms already record. The platform builds a model of what each individual turbine should produce in the conditions it actually experiences, measures the real machine against it continuously, and reports the difference as a ranked set of findings with the annual energy and revenue attached to each one.

Wind

The only sector we serve

SCADA

The only required input

On + offshore

Fleets covered

Per turbine

The unit of analysis

A performance model for every turbine, not another dashboard.

Most software sold to wind operators is a reporting layer. It renders data the operator already held, more attractively, and leaves the interpretation to the engineering team. That is a reasonable business, and it places a hard ceiling on how much production it can recover, because the constraint was never the chart.

OpenTurbine begins at the other end. The platform builds a machine learning model of what each individual turbine should produce in the conditions it actually experienced, then measures the real machine against that expectation continuously. A loss becomes a named turbine with a probable cause and a figure attached, rather than a line on a report that is slightly lower than last year.

The commercial consequence is that it changes what an operator can act on. A ranked list of recoverable losses is a maintenance plan. A dashboard is an additional analysis task for a team that already has one.

How a per-turbine expectation is built

The lowest-cost clean energy is the production already paid for.

The installed base is now large enough that improving it matters more than adding to it. Fleets built over the past fifteen years are producing measurably less than they could through drift, wear, and settings that have not been revisited, and none of that is visible without a per-turbine expectation to measure against.

Recovering it is one of the few cases in which the sustainable decision and the profitable decision are the same decision. More energy from the same asset is more revenue and more clean generation at once, with no new turbines, no additional land, and no capital project to approve.

That is the premise of the company. Everything else - the modeling, the fault detection, the value attached to every finding - is what it takes to make that recovery specific enough for an operations team to act on.

How we work with an operating fleet.

Every engagement starts with historical data rather than with a deployment. An operator can establish whether the models read their fleet correctly before anything is connected, and that sequence is deliberate: a vendor that requires installation before it will produce evidence has asked the customer to carry the risk of the evaluation.

Findings are delivered with their evidence rather than as scores. Each one names the turbine, the period, the signals that moved, and the expectation the machine was measured against, because an operations team is accountable for the intervention and cannot defend a work order that rests on a number with no derivation.

We report negative results as readily as positive ones. If an assessment shows a fleet is operating close to its achievable output, that is the finding, and saying so is worth more over the life of a relationship than a list of marginal recommendations.

What the first engagement involves, stage by stage

The principles the product is built on.

The team comes from machine learning, data engineering, and wind energy engineering. These four principles are the ones that decide design questions when those three disciplines disagree.

Measured, not asserted

Every claim the platform makes about a turbine is traceable to the data that produced it. A finding an engineer cannot check is not evidence.

Existing assets first

The installed base is large enough that improving it matters more than adding to it. Recovered production requires no capital project and no additional land.

Precision over coverage

A false positive commits a crew, and offshore it commits a vessel. Reporting fewer findings with the evidence attached is worth more than reporting everything.

Operations, not analytics

The deliverable is a sequenced work list with a figure against each item, because that is what an operations team can act on and defend afterwards.

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.

Request a demo Contact us