top of page

How Turbit AI detects minimal power losses early on

As the world becomes more dependent on renewable energy, wind turbines have become a common sight across the globe. However, maintaining these turbines and ensuring their efficient operation can be a challenging task. That's where Turbit comes in. Turbit is a software company that specializes in developing and deploying AI monitoring at scale for wind parks. In this article, we'll dive into what Turbit does and how it can help prevent unnoticed power anomalies or main component breakdowns.

To begin with, Turbit's AI technology is trained to predict the power output of a wind turbine based on various factors. The network is trained with historical data of that particular turbine at its specific location, which includes outside temperature, wind speed, wind direction, turbulence intensity, and the position of the turbine within the wind park. This means that Turbit can learn the power output of the turbine very accurately, even accounting for sound reduction curves during the night and specific behaviors that occur only from one direction.

The result of Turbit's AI technology is a much more precise measurement of the power output of the turbine. This level of precision can be up to plus or minus 30 kilowatts. That means fewer false alarms and the ability to detect even the slightest deviations from normal behavior. For instance, after a power curve update or a turbine control system upgrade, Turbit can measure if something is going better or worse. This level of precision is achieved by analyzing every new 10-minute average value, and the whole system runs on just SCADA data, which is readily available.

In one specific example, Turbit detected a very slight deviation of about 100 to 200 kilowatts of difference, and sent an alarm to the customer. Upon investigation, it was found that the pitch system of the blades was not working correctly, which led to a pitch offset of six degrees and reduced energy production. The manufacturer was contacted, and it was discovered that they had forgotten to reset the pitch. The problem was solved after just three days, preventing a power loss of over 20,000 euros that would have otherwise gone undetected for weeks or months.

This example illustrates the value of Turbit's system, which automatically detects failures and eliminates the need for manual work. Operators and asset managers can act as early as possible on abnormal wind turbine behavior, plan repairs months ahead of failures, and monitor repair quality. This feedback loop from users constantly improves detection accuracy and increases communication relevance from Turbit's AI.

In conclusion, Turbit is a software company that provides AI monitoring at scale for wind parks. Its technology enables operators and asset managers to act quickly and efficiently to prevent unnoticed power anomalies or main component breakdowns. With Turbit, wind park operators can plan repairs months ahead of failures and monitor repair quality, thus saving time and money while ensuring reliable and efficient energy production.