Accounting for Environmental Conditions in Data-Driven Wind Turbine Power Models

Continuous assessment of wind turbine performance is a key to maximising power generation at very low cost. A curve non-linear function between output and speed widely used approach numerous problems linked operation. According the current IEC standard, curves are determined by data reduction method, called binning, where hub height, air density considered as appropriate input parameters. However, rotors have grown in size over recent years, impact variations speed, thus output, can no longer be overlooked. Two environmental variables, namely shear turbulence intensity, greatest on output. Therefore, taking account these factors may improve accuracy well reduce uncertainty data-driven models, which could helpful monitoring applications. This paper aims quantify analyse two curves. Gaussian process (GP) data-driven, nonparametric based modelling that incorporate additional factors. The proposed technique's effectiveness trained validated using historical 10-minute average supervisory control acquisition (SCADA) datasets from variable pitch control, turbines rated 2.5 MW. results suggest (i) inclusion parameters increases GP model reduces estimating curve; (ii) comparative study reveals intensity has relatively greater accuracy, together with compared blade angle. These conclusions confirmed error metrics calculations. practical beneficial consequences for O&M related activities such early failure detection.