Marine predators algorithm with deep learning based solar photovoltaic system modelling and optimization of green hydrogen production

Green hydrogen production depends on solar energy contains employing renewable solar power to purpose the electrolysis of water, causing the separation of hydrogen and oxygen molecules. This method contains a high potential to address the energy transition challenges caused by weather changes and vital carbon-neutral alternatives. Photovoltaic (PV) based combined energy systems performance as a potential new technological solution for clean and affordable green hydrogen production. Optimizing solar PV systems for the effectual generation of green hydrogen includes maximizing the energy output of solar panels and increasing the entire hydrogen production method. Several researchers and scientists are paid attention to the optimizer and modelling of many blocks developing the PV electrolysis method to acquire the optimum solution. With this motivation, the study presents a new Marine Predator Algorithm with Deep Learning-based Modelling and Optimization of the Green Hydrogen Production (MPADL-MOGHP) technique. The MPADL-MOGHP technique can be employed for determining the optimum operational variables of the water electrolysis procedure relevant to hydrogen (H2) production. Catalyst amount (μg), electrolysis time (min), and electric voltage (V) are the three controlling factors that should be properly detected to increase hydrogen production. The MPADL-MOGHP technique comprises two major stages of operations such as modelling and optimization. Primarily, the DBN model is applied for simulating the water electrolysis procedure designed based on electric voltage, quantity of catalyst, and electrolysis time. Next, the MPA is applied for determining the optimum parameters of the water electrolysis process to maximize the generation rate of the hydrogen. The quantity of catalyst, the electrolysis time, and the electric voltage are applied as decision parameters during the optimization process. The performances of the MPADL-MOGHP system are tested on different aspects. The experimental values highlighted the promising results of the MPADL-MOGHP method over other existing techniques.