Harmonics Prediction and Mitigation using Adaptive Neuro Fuzzy Inference System Model based on Hybrid of Wind Solar Driven by DFIG

In this paper a filter design methodology is proposed based on renewable energy hybrid model. The main function of the designed filter is to reduce the harmonics in distributed power systems consisting of hybrid renewable resources, wind and solar, leading to an increase in efficiency of such systems. The existence of harmonics in these systems is a direct result of the energy fluctuations and intermittency of the renewable resources. This necessitates the use of filtering system to mitigate the effect of these harmonics. The proposed model consists of wind turbine driven by Doubly Fed Induction Generator (DFIG) and a solar photovoltaic (PV). It is shown through the hybrid model that the generated power is not stable due to the wind speed and solar irradiation fluctuation. The forecasting phase of this work comes after establishing the hybrid model and generating the electrical power output. In this phase a model is developed using the simulated data and other relevant parameters as input for training to predict an hour and a day ahead. This work uses the Adaptive Neuro Fuzzy Inference System (ANFIS) for forecasting purposes. Finally, the data obtained from the hybrid model simulations are analyzed in order to build an understand of the behavior. The inputs from study and forecast serve as inputs to select from a combination of filters to effectively mitigate harmonics for the proposed system.