The focus of this presentation will be to share the latest results of a multi-year DOE funded systems analysis program dedicated to investigations of hydrogen production infrastructures. The project goal is to develop a better understanding of how the production infrastructure will develop during the transition to a hydrogen economy in the 2015 timeframe and beyond. The main tool used in developing this understanding is HyPro, a MATLAB computer simulation of the hydrogen production facility build-out over time. The simulation is in its second generation which includes more accurate costing models and provides the user with more in-depth analysis. This simulation reads cost and performance data of hydrogen production, delivery and dispensing and performs a discounted cash flow analysis of the various options under a set of economic and demand conditions. The inputs or data were collected from a literature review and consultation with industry. As new information arises the inputs are verified and improved to determine the impact on the simulation.

The model is capable of evaluating over 100 hydrogen pathways. These pathways consist of various hydrogen production, delivery and dispensing methods. Among the production options are regional, central, city gate, and forecourt size plants using different feedstocks such as natural gas, coal, biomass, and water in steam methane reforming, gasification, and electrolysis processes. For delivery, different trucking options, both gaseous and liquid, as well as pipelines are evaluated. Each segment of the infrastructure is combined with a compatible downstream segment to create the pathways analyzed by the model. The basic premise for the model is that the demand must always be met. The demand and supply is evaluated each year and if there is a deficit the lowest cost pathway is selected and built. The analysis results will be described in several ways. The model provides the pump cost of hydrogen in $/kg for each of the pathways. Additional results include the lowest cost pathway, total infrastructure costs, stranded assets and segment costs.

The model is flexible enough to allow for different types of scenarios to be postulated. Among the results to be presented will be the many economic and demand conditions that have been evaluated. The impact of input data error bounds on the program results will be shared. Externalities like policies and emissions requirements can have a significant influence in the form that the infrastructure takes. Case studies showing the effect of zero emission policies on the build-out structure will be described. Lastly, the impact of other external variables such as feedstock costs, capital cost credits, and higher risk production methods will be presented.