Intelligent Wind Turbine Power Curve Modelling Using the Third Version of Cultural Algorithm (CA3)

Abstract

third version of cultural algorithm (CA3), error analysis, mathematical modelling, quadratic Gaussian function, wind turbine generator power curve (WTGPC).The wind turbine generator power curve (WTGPC) gives the relationship between the wind speed and power output of the wind turbine at any given time. The power curves, which are usually provided by the manufacturer company, are mainly used in forecasting, energy planning and performance monitoring of wind turbines. The WTGPC model plays a significant role in the control and monitoring of wind farms as well as playing a role in the wind farms power injection to the grid. This paper presents a comprehensive analysis of several methods of modelling the WTGPC, with respect to four commercial wind turbines; 330, 900, 2000 and 3050 kW. In the first step, the proposed method of the study, based on quadratic Gaussian function, is compared to several developed mathematical models by using error measurement techniques including mean square error (MSE) and residual analysis. The accuracy of the proposed method has then been improved by means of the third version of cultural algorithm (CA3) through the optimization of the proposed method coefficients. The ultimate performance of the compared methods has been investigated by the normalized root mean squared error (NRMSE), where the proposed method of the study shows an excellent performance for modelling of wind turbine power curves.