Abstract
This research is dedicated to designing and economically evaluating the green ammonia supply chain, considering the fluctuating nature of renewable energy sources and energy demand across both hourly and seasonal variations. It also explores the impact of economies of scale and the delays associated with long-distance shipping to meet energy demands in a timely manner. These considerations require the formulation of a Mixed-Integer Nonlinear Programming model, further complicated by the necessity for a two-stage stochastic programming approach. We introduce a hierarchical optimization framework that utilizes a decomposition method to differentiate between one-time design decisions and subsequent operational choices. At the upper level, potential design solutions are identified through the Bayesian Optimization and Hyperband algorithm, which effectively navigates the non-linear challenges posed by economies of scale. The lower level then addresses a Mixed-Integer Linear Programming problem to independently assess the feasibility of each scenario. Our empirical analysis includes case studies of three potential international routes for transporting green ammonia to Korea. We contrasted our methodology with a hypothetical scenario that presupposes a constant supply of power and a stable demand for energy. Additionally, techno-economic analyses were conducted to evaluate the implications of the minimum operational limits for electrolyzers.