Fuzzy based prediction of wind distributed generation impact on distribution network: Case study—Banat region, Serbia

Abstract

The paper proposes the methodology for the assessment of impact that wind distributed generation can have on distribution network. Effects of active power losses, voltage drops, and voltage total harmonic distortion are considered. The methodology uses fuzzy logic in order to address uncertainties in wind energy generation, as well as artificial neural networks for wind speed assessment. Measurement data of temperature, irradiation, and wind direction at different wind turbine hub heights are used. The predicted wind speed, active power loss variation, and elasticity of power quality constraints are presented in a form of fuzzy numbers. The proposed methodology is tested on realistic, 28-bus 35 kV voltage distribution system in the Banat region in Serbia. A steady-state voltage stability index is applied to select wind distribution generation locations. It is shown that wind distributed generation can improve voltage regulation and reduce active power losses. However, increase in voltage total harmonic distortion is possible. The proposed methodology enables analysis of the impact that wind speed has on possible effects of wind distributed generation.