Neural network-based optimization of hydrogen fuel production energy system with proton exchange electrolyzer supported nanomaterial

• Published in: Fuel (Guildford) Vol. 332; p. 125827
• Main Authors: Hai, Tao, Hikmat Hama Aziz, Kosar, Zhou, Jincheng, Dhahad, Hayder A., Sharma, Kamal, Fahad Almojil, Sattam, Ibrahim Almohana, Abdulaziz, Fahmi Alali, Abdulrhman, Ismail Kh, Teeba, Mehrez, Sadok, Abdelrahman, Anas
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.01.2023
• Subjects: Biofuel; Electrolyzer; Fuel production; Hydrogen fuel; Optimization; Solar energy; Fuel production; Biofuel; Hydrogen fuel; Electrolyzer; Optimization; Solar energy
• ISSN: 0016-2361; 1873-7153
• DOI: 10.1016/j.fuel.2022.125827

•Simulation of a hydrogen energy system.•Using PEM electrolysis for the hydrogen productionon.•Techno-economic assessment of the system for the feasibility study.•Multi-objective genetic algorithm optimization.•Neural network-based optimization of the proposed system. In this research, we try to investigate a solar-geothermal energy system. This system includes three turbines for power production, a PEM electrolyzer for hydrogen production, and a thermoelectric for generating electricity from excess heat. In addition, the seawater will be passed through the osmotic cycle to gain fresh water. The required power for this osmotic cycle will be obtained through the energy produced by the main turbines. The generated load, hydrogen production flow rate, purified water flow rate, and heating consumption are assessed in this study. The results showed that this system can produce 3.8 megawatts of electricity as well as 8 g per second of hydrogen fuel at the operating point. Also, the energy efficiency of this system is estimated to be 19%. Afterward, machine learning methods are used to optimize designing parameters, and the optimum operating point in terms of useful power and stored fuel flow rate is obtained by a genetic algorithm. The optimum operating point of this energy system has a useful power output of 4.099 megawatts and a hydrogen flow rate of 29 g per second. In the end, the distribution of the design parameters is displayed for points of the beam curve.