Abstract (EN):
This paper evaluates the results that can be obtained with different strategies of instrumentation for an utility-scale wind turbine, considering variations on the number of sensors, on their distribution in the wind turbine tower and on their noise level. With the goal of reducing the investment in monitoring equipment, layouts based on a reduced number of low cost sensors are tested. The data processing strategy is based on the continuous tracking of the wind turbine main vibration modes using Operational Modal Analysis techniques combined with algorithms that deal with the particularities of operation of the wind turbines and that permit an online automated identification. After the mitigation of the influence of the environmental and operation factors on the tracked natural frequencies, it is possible to detect abnormal variations of the natural frequencies, which might flag the appearance of damage. The alternative monitoring scenarios are recreated using a database of acceleration time series collected during one year in an onshore 2MW wind turbine, by a quite extensive monitoring system, using low noise accelerometers.
Language:
English
Type (Professor's evaluation):
Scientific
No. of pages:
8