Energy management in hybrid photovoltaic-wind system using optimized neural network

In recent era, the reduction of greenhouse gas emission and fuel consumption is accompanied by adopting photovoltaic (PV) and wind turbine-based hybrid renewable energy sources (HRES). In nature, an intermittent characteristic of the wind speed and solar irradiation makes these sources unpredictable, and hence, energy produced by wind and PV system generates uncertain conditions in operation of microgrid. In such cases, the security and reliability of microgrid are enhanced by integration of energy storage system (ESS). This work deals with an energy management in a hybrid system incorporating PV source and permanent magnet synchronous generator (PMSG)-based wind energy system. The PV output is enhanced with a help of switched trans-quasi-Z-source (TQZS) boost converter in which cuckoo search-assisted radial basis function neural network (RBFNN) approach is used as maximum power point tracking (MPPT) technique for tracking maximum photovoltaic power. The proposed approach results in high-power tracking efficiency with reduced power loss and settling time. A battery is incorporated to address an intermittent nature of RES. Artificial neural networks (ANN), which are capable of self-learning battery dynamics, keep track of the state-of-charge (SOC) of the battery. The system thus framed is implemented using MATLAB software, and promising results are generated in terms of power management with improved efficiency of 92%, gain ratio of 1:10 and total harmonic distortion (THD) value of 2.33%, respectively.