Multi-objective optimisation of a hydrogen supply chain

Hydrogen produced from renewable sources and used in fuel cells both for mobile and stationary applications constitutes a very promising energy carrier in a context of sustainable development. Yet the strategic roadmaps that were currently published about the energy potentialities of hydrogen at European, national and regional level as well as the analysis of the scientific publications in this field have identified the lack of infrastructures as a major barrier to the development of a « hydrogen » economy. This study focuses on the development of a methodological framework for the design of a hydrogen supply chain (HSC) (production, storage and transportation). The formulation based on mixed integer linear programming involves a multi-criteria approach where three objectives have to be optimised simultaneously, i.e., cost, global warming potential and safety risk, either at national or regional scale. This problem is solved by implementing lexicographic and Ɛ-constraint methods. The solution consists of a Pareto front, corresponding to different design strategies in the associated variable space. Multiple choice decision making based on M-TOPSIS (Modified Technique for Order Preference by Similarity to Ideal Solution) analysis is then selected to find the best compromise. The mathematical model is applied to a case study reported in the literature survey and dedicated to Great Britain for validation purpose, comparing the results between mono- and multi-objective approaches. In the regional case, the modelling and optimisation of the HSC in the Midi-Pyrénées region was carried out in the framework of the project “H2 as a green fuel”. A mono/multi period problem is treated with different optimisation scenarios using Ɛ-constraint and lexicographic methods for the optimisation stage. The geographic information system (GIS) is introduced and allows organising, analysing and mapping spatial data. The optimisation of the HSC is then applied to the national case of France. The objective is twofold: on the one hand, to examine if the methodology is robust enough to tackle a different geographic scale and second to see if the regional approach is consistent with the national scale. In this case study, the ArcGIS® spatial tool is used before optimisation to identify the geographic items that are further used in the optimisation step. A scenario with an economic cycle is also considered. Mono- and multi-objective optimisations exhibit some differences concerning the degree of centralisation of the network and the selection of the production technology type and size. The obtained results confirm that different spatial and temporal scales are required to encompass the complexity of the problem