Greenbyte Energy Cloud explores the symptoms of failing wind turbines with AI
Predict, the ambitious technological leap in Greenbyte Energy Cloud, is a new digital condition monitoring feature that uses artificial intelligence and SCADA data, to identify failures in wind turbine components before they occur.
Greenbyte’s focus on the needs of the renewable energy industry professionals shed a light to the effects of wind turbine shortcomings. Component failures hinder efficiency and negatively affect budget targets for wind farm operators and owners. However, this is an everyday occurrence for operators, forced onto unplanned downtime and immediate maintenance, resulting to loss of revenue, discrepancy of planned budget and unnecessary time loss. Mikael Baros, Director of Technology at Greenbyte, comments accordingly in his blog post for Predict:
We expect turbines to operate 24 hours a day, 7 days a week. If we did the same with a car, it would only last us 8 months! Hence it is not surprising that these poor turbines fail (too) often. It is estimated that up to 30% of the total life-cycle cost of a wind farm is due to failure and maintenance activities.
Predict aims to target this problem by using machine learning to teach the system how to alarm on possible upcoming failures in the frailest wind turbine components. The sophisticated module uses SCADA data to estimate the expected temperature for critical components, compares the estimated data to the actual measured values, and enables intelligent and early detection of developing failures. The pilot study on Predict detected faults 2 to 9 months in advance, and has presently achieved 94% accuracy and 23% reduction of cost. Using machine learning, the software keeps improving accuracy, as the more data is accumulated, the more solid the predictions can be.
In Greenbyte’s new technology blog, Director of Technology, Mikael Baros shares the compelling story of how Predict came to be a reality through AI; The feature learned how to crunch real time data and form insights on the state of components by guiding machine learning towards these particular needs. Read the full article here.