A methodology of a hybrid hydrogen supply network (HHSN) under alternative energy resources (AERs) of hydrogen footprint constraint for sustainable energy production (SEP)

We aim to suggest a methodology of a smart hybrid hydrogen supply network based on diverse alternative energy resources of hydrogen footprint constraint. To date, hydrogen production has been mostly dependent on fossil fuels. However, future hydrogen would be harnessed contingent on eco-friendly energy resources to support environmentally benign hydrogen economy. In this study, a smart hybrid hydrogen supply network is designed considering hydrogen production from solar energy, wind energy and wastewater and hydrogen distribution by using reinforcement learning. A mathematical model is divided into two phases. First phase is a stochastic programming under demand uncertainty, where multi objective functions are to minimize the total annual costs and environmental costs, respectively. Second phase is a heuristic optimization problem based on Q-learning which is one of the reinforcement learning algorithms. The suggested model is applied to Gyeongsang province in the Republic of Korea as a case study. Alternative energy resources are selected considering regional characteristics. We verify possibilities for construction of a smart future hydrogen supply network based on various feasible scenarios, where can propose the best hydrogen network to decision-makers.