Wind Power Curve Estimation by Adaptive Neuro Fuzzy Inference System

Abstract

Sensitive forecasting of wind turbine output power curve has a significant role in tracing of conditioning and controlling of wind turbines
in wind power plants. This study involves the adaptive neuro fuzzy inference system (ANFIS) in forecasting of horizontal axis wind turbine
output power (P) in terms of climatic data and turbine rotational speed (Ω). Adaptive neuro fuzzy inference system (ANFIS) is based on the
input parameters covering the wind speed (V), atmospheric air temperature (T) and the wind turbine rotational speed (Ω). These parameters
were obtained from an operating power plant in Turkey. Derived results of the testing process reveal that minimum mean absolute percentage
of error (MAPE) and maximum correlation coefficient (R) values were determined for an optimum rotational speed (Ω). MAPE and R values
were evaluated to be 1.18% and 0.9992, respectively when ANFIS approach was applied to the wind farm data. These results indicated well that
ANFIS approach presented a basic and precise estimation in wind turbine output power (P) determination. A wind farm’s turbine power curve
can be quickly forecasted in an accomplished way under little error through the utilization of the ANFIS method when the parameters of wind
speed (V), atmospheric air temperature (T) and rotational speed (Ω) of wind turbines in a wind farm are known. Thus, this technique is rather
suitable in decision making process of new wind power plant installations of a considered zone.