Abstract
Wind turbines are rotating machines which are subjected to non-stationary conditions and their power depends non-trivially on ambient conditions and working parameters. Therefore, monitoring the performance of wind turbines is a complicated task because it is critical to construct normal behavior models for the theoretical power which should be extracted. The power curve is the relation between the wind speed and the power and it is widely used to monitor wind turbine performance. Nowadays, it is commonly accepted that a reliable model for the power curve should be customized on the wind turbine and on the site of interest: this has boosted the use of SCADA for data-driven approaches to wind turbine power curve and has therefore stimulated the use of artificial intelligence and applied statistics methods. In this regard, a promising line of research regards multivariate approaches to the wind turbine power curve: these are based on incorporating additional environmental information or working parameters as input variables for the data-driven model, whose output is the produced power. The rationale for a multivariate approach to wind turbine power curve is the potential decrease of the error metrics of the regression: this allows monitoring the performance of the target wind turbine more precisely. On these grounds, in this manuscript, the state-of-the-art is discussed as regards multivariate SCADA data analysis methods for wind turbine power curve modeling and some promising research perspectives are indicated.
Highlights
Modern horizontal-axis wind turbines are complex machines which are subjected to non-stationary conditions; monitoring their performance is a complicated task which has attracted a wide debate in the scientific literature.Typically, the manufacturer of a wind turbine provides standards for the behavior of the machine, basing on field test in a controlled environment: these are expressed in the form of curves for the thrust coefficient and for the power coefficient
The power curve of a wind turbine is fundamental to understand its performance because it is given by the measured relation between the wind speed and the output power, which can be visualized through a simple two-dimensional scatter plot
It should be noticed that these error metrics, obtained in a preliminary study, are lower with respect to all the results reported in Section 2 for models employing only SCADA data, because the MAE is approximately 0.5% of the rated power and the AEP is estimated with a precision in the order of 0.3%: this supports the usefulness of this approach to multivariate wind turbine power curve modeling
Summary
Modern horizontal-axis wind turbines are complex machines which are subjected to non-stationary conditions; monitoring their performance is a complicated task which has attracted a wide debate in the scientific literature. On the grounds of Equation (3) and Figures 3–7, it arises intuitively that knowing, for example, the blade pitch and the rotor speed in addition to the wind speed could be helpful for predicting how much power the wind turbine should extract This means that the normal behavior model for the performance of a wind turbine should preferably employ more than one input variable (wind speed). To [17] and differently with respect to the standard in the literature (see Section 2), the minimum, maximum and standard deviation of the main measurements are included as possible covariates, in addition to the average values
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call