Abstract
The power curve of a wind turbine plays an essential role from its design till the end of its life. The production of energy depends on the relation power/wind-speed defined by the power curve. Any deviation of the expected relation between these two variables has an important impact translated usually in money for the owner in different ways. If the power curve observed over a period of time is different from that available in a previous interval of time analysed, usually effects of faults or natural or induced degradations are present. This paper describes how to reduce the huge amount of information based on collecting in real-time pairs of values power/wind-speed to few essential pairs able to describe the whole power curve by a model. Using this model it is possible to define a power curve reference in order to compare with other future performances. Also, this will allow for the detection of anomalies. The methods used for those illustrated by examples are the main contribution included in this paper.
