Economic assessment of efficient hydrogen production-based hybrid renewable energy system: OOA-RBFNN approach

Abstract

A sustainable society is thought to be greatly aided by hydrogen (H₂) energy as it is a clean and efficient energy source in light of the impending energy revolution and global climate change. Identifying and implementing green H₂ production methods is made considerably more difficult by the need for a gradual switch to renewable energy. To address these issues, this study proposes a novel energy management approach for hybrid renewable energy resources (RES) systems using multiple H₂ production methods. The proposed approach combines the osprey optimization algorithm (OOA) with a radial basis function neural network (RBFNN), known as the OOA-RBFNN technique. The principal purpose of the proposed strategy is to minimize net system costs. Specifically, OOA is used to lessen the operational cost of a hybrid microgrid consisting of RES. RBFNN is used to predict uncertain renewable energy generation and demand. This work aims to present a strategy for producing hydrogen from solar and wind energy while reducing system costs by using water electrolyzer. The OOA-RBFNN technique is used to define the optimal size and operating energy management of the system. The proposed technique was implemented in the MATLAB platform and compared with various existing techniques like the salp swarm algorithm, convolutional neural network and random forest algorithm. The computation time of the proposed approach is 0.8 s which is lower, and the cost for energy is 23.22$ which is lower than the existing methods.