Set Up Data Science Pipelines at Scale
Technical Set Up
For each turbine, component and each site, Turbit AI learns the normal performance behavior from historical SCADA data.
Turbit models use physically relevant input data such as wind speed, temperature, wind direction and turbulence intensity for each turbine and component.
The machine learning models predict the performance with an accuracy of over 99%.
Turbit AI in Production
Turbit compares constantly simulated with actual data. The algorithms detect the performance anomaly of each turbine independently regardless of the manufacturer and site.
Detected anomalies are compared with our database of over 6000 turbine years and classified in availability categories. Furthermore, we compare each event with the turbine status to add more context to the information and filter unimportant events.
Turbit sends relevant insights like yield potentials or an unnoticed temperature rise in the generator in a real-time data analysis via e-mail. Within Turbit Web App there is even more relevant information.