AI-Enabled Prediction of Wind Speed and Temperature with Multiverse Optimization for Efficient Control of Standalone Hybrid System

Instead of going for one renewable source for improved efficiency and stability, hybridization of energy sources is demonstrated. The main objective of the work is to do analysis of different perspectives of Hybrid Renewable Energy Storage Systems (HRESS) like, power architectures, mathematical modelling and power electronic converter topology and design optimization algorithms. Non-conventional energy sources like solar and wind play the vital role for improved energy production. Different power converters with proper control strategies are illustrated to maintain the Direct Current (DC) link voltage irrespective of any dynamic changes. For achieving minimum fluctuations from the output of renewable energy conversion system wind speed and temperature forecasting are done with machine learning. Using various regression techniques, with the prediction of wind speed and temperature the model accuracy is improved. Uninterrupted power supply is ensured by using battery. Maximum power extraction from Photo Voltaic (PV) is taken care by Maximum Power Point Tracking (MPPT) algorithm. Wind Energy Conversion System (WECS) works at its maximum efficiency with Permanent Magnet Synchronous Generator (PMSG) coupled to the wind turbine (WT). For proper power management power converters like boost converter and bidirectional DC-DC converter are controlled with Proportional Integral (PI) controllers. For controlling the parameters of PI along with conventional techniques, nature inspired optimization techniques are also implemented. The comparative analysis is also done with various algorithms. LC filters are installed to ripple out the unwanted components from the output voltage. The simulation of the model is verified using MATLAB/Simulink and steady state performance is analyzed.