Hybrid photovoltaic wind renewable energy sources for microgrid using osprey optimization algorithm and augmented physics informed neural network
K. E. Lakshmi prabha, R Deepa, G. T. Shakila Devi, K. S. Balamurugan
Abstract
The integration of hybrid photovoltaic (PV) and wind turbine (WT) renewable energy sources (RESs) into a microgrid (MG) offers a considerable opportunity for clean and cost-effective energy solutions, but optimization of the trade-off between generation and usage of energy is a major challenge. To overcome these challenges, this paper proposes a hybrid technique for efficient incorporation and management of hybrid PV and WTRESs in MGs, ensuring improved energy stability, reliability, and cost-effectiveness. The suggested method integrates the osprey optimization algorithm (OOA) and augmented physics-informed neural network (APINN); hence, it is called the “OOA-APINN” technique. The goal is to increase the economic performance of hybrid PV-wind RES by lowering the cost of energy (COE). The OOA is employed to optimize the operational parameters of the PV-wind system and the integrated energy storage, ensuring efficient energy management (EM) and cost-effective operation. The APINN is employed to forecast energy generation and consumption trends, enabling optimized scheduling as well as resource allocation for improved EM. By then, the suggested technique is executed on the MATLAB platform and evaluated with various existing approaches, such as coati optimization algorithm, fuzzy decision maker based multi-objective optimization algorithm, multi-objective particle swarm optimization, robust optimization, and gray wolf cuckoo search algorithm. The suggested OOA-APINN method achieves the lowest COE at $0.16/kWh, demonstrating a substantial enhancement in cost effectiveness compared to other optimization methods.