Regression model for predicting the speed of wind flows for energy needs based on fuzzy logic.

Renewable energy integration becomes an important criterion for the sustainable generation of electrical power. The high introduction of wind power plants into the power system leads to some inconveniences in the power system operators' work due to the unpredictable and variable nature of the wind speed and the power generated by wind farms. Even though the power generated at the wind power plants is not regulated by the system operator, the accurate predicting of the wind speed and the angle of its direction could solve such problems leading to improving the reliability of power supply systems. This paper considers the prediction of both the speed and direction of wind flows for the Far East coast. It is proposed to use autoregression based on the fuzzy systems concept. The goal of this study is to find a regression model that satisfies all observable fuzzy data within the specified optimality criterion. Thus, the regression coefficients are fuzzy numbers that can be expressed as interval numbers with membership values. As result, the electric power generated by wind power plants could provide additional energy accumulation function, if the wind flow velocity fluctuations will be considered. • A novel predicting method of wind speed was proposed based on a fuzzy regression. • The developed method allows assessing the maximum power output of a wind turbine. • The obtained computation results showed acceptable accuracy 3 h ahead. • The consideration of the wind flow velocity fluctuations provides additional energy accumulation in the wind power plants. • The proposed method allows to increase the reliability of the power supply. [ABSTRACT FROM AUTHOR]