Employing Crocodile Hunting Search Approaches to Manage Renewable Energy-Based Microgrid

Abstract

This study explores the benefits and achievements of Machine Learning models in guiding active optimization procedures for selecting the most appropriate hybrid renewable energy system. The system is known for its cost-effectiveness, reliability, and performance. The study focused on a specific example of a hybrid renewable energy system that combines solar photovoltaic and wind resources. The study also considered the use of a fuel cell to store any extra electricity generated. This work aimed to conduct a crocodile hunting search, which was achieved using optimization models. The MATLAB program and Simulink environment were utilized to implement and execute numerical simulation processes. Based on the numerical simulations, it was found that the hybrid renewable energy system consumed a total of 19.88 grams of hydrogen fuel. Meanwhile, the poorest performance was observed in CHS, which exhibited the highest hydrogen fuel usage, amounting to 25.73 g. The fuel cell voltage for CHS exhibited a range of 43 to 49.5 V. Furthermore, it is customary for the fuel cell current to fluctuate between around 20 and 160 A. However, the CHS model exhibited a wider range of fuel cell currents, ranging from 75 to 210 A. The average range for hydrogen fuel usage varied between approximately 10 and 75 L/m. In addition, it was discovered that all algorithms measured a hydrogen fuel consumption rate of 20 g, except for the CHS paradigm, which achieved a maximum hydrogen fuel consumption rate of 25 g. The whole duration of the simulation in this study was 350 seconds.