Integration of Wind and PV Systems Using Genetic-Assisted Artificial Neural Network

Abstract

The prominence of Renewable Energy Sources (RES) in the process of power generation is exponentially increased in the recent days since these sources assist in minimizing the environmental contamination. A grid-tied DFIG (Doubly Fed Induction Generator) based WECS (Wind Energy Conversion System) is introduced in this work, in which a Landsman converter is implemented to improvise the output voltage of PV without any fluctuations. A novel GA (Genetic Algorithm) assisted ANN (Artificial Neural Network) is employed for tracking the Maximum power from PV. Among the rotor and grid side controllers, the former is implemented by combining the stator flux with d-q reference frame and the latter is realized by the PI controller. The proposed approach delivers better performance in the compensation of real and reactive power along with the DC link voltage control. The controlling mechanism is verified in both MATLAB and experimental bench setupby using an emulated wind turbine for the concurrent control of DC link potential, active and reactive powers.The source current THD is observed as 1.93% and 2.4% for simulation and hardware implementation respectively.