May 26, 2020 AUTOMOTIVE, WHITEPAPERS
Can you detect a 15 cm crack in a wind turbine blade without stopping the wind turbine? A three-year-long research project, partly financed by the Danish government, proved that this is indeed possible.
Modern wind turbine blades are designed to last for 20 to 25 years under severe weather conditions, and during this period, damage is unavoidable. Almost inevitably, a small blade defect will develop into a bigger failure, which if no countermeasures are taken, will become critical, causing serious consequences.
Repairing a small defect is significantly cheaper than repairing a bigger one or replacing an entire blade. Therefore, wind turbine operators pay close attention to monitoring their fleets’ blades. Today, this is done by periodical visual inspections conducted every one to two years. For such an inspection, the wind turbine has to be stopped, then a service technician, often trained climbers, checks every inch of the blade’s surface and documents any defects found. This is a tedious and risky job that can only be done in good weather conditions. Since such inspections are quite expensive, many in the industry realize that a better approach is needed.
One solution could be an automatic, remote, structural health monitoring system that can instantaneously identify damage without stopping the wind turbine, thus avoiding expensive downtime. Brüel & Kjær, in cooperation with leading Danish wind turbine specialists, has developed a prototype of such a system – an active, vibration-based, structural health monitoring system that uses an actuator and accelerometers to collect vibration data, while a damage detection algorithm indicates if blade damage is present. The solution was used to monitor a real Vestas V27 wind turbine for four months, resulting in the detection of a 15 cm crack on a blade’s trailing edge.
Not just for wind turbines
It is believed that the same principle can be used for monitoring pipelines, tanks, chimneys, aircraft fuselages, and other thin-shell structures.