Global techno-economic analysis of inter-annual energy storage in 100% renewable energy systems with comparative assessment of multi-objective solutions, e-methane versus e-hydrogen, and data series length

Abstract

The global shift to 100% renewable energy systems is vital for climate action. However, guaranteeing year-round reliability depends on large-scale inter-annual storage to manage the natural fluctuations of solar and wind power. This thesis provides a detailed global techno-economic analysis of inter-annual storage needs, covering 145 global regions across the power, heating, transportation, and desalination sectors. The research tackles key design challenges, namely the selection of storage technologies and the significant effect of long-term weather patterns. It moves beyond a one-size-fits-all approach, showing that the best solutions are unique to each region and depend on a crucial balance between building extra renewable generation capacity and investing in storage. This research is the first global assessment to systematically measure the trade-offs between generation overcapacity and storage infrastructure, creating a foundational guide for designing resilient, and sustainable energy systems.

Initial analysis across 145 global regions using NASA weather data from 1984-2005, based on an electricity-based hydrogen storage system, reveals two distinct pathways for system design. A cost-optimised approach requires a 5.0% overcapacity in renewable electricity generation with no additional inter-annual storage, adding a modest 3.3% to the baseline cost of a 100% renewable energy system in 2050. In contrast, a curtailment-optimised scenario uses 1.4% overcapacity but demands storage volumes of 417.4 TWhH₂,LHV of hydrogen, 0.8 TWhCH₄,LHV of methane, and 4.2 TWhth,LHV of liquid fuels, which raises the baseline system cost by 103.1%. Further investigation confirms that for inter-annual balancing, electricity-based methane is the superior economic choice, with an annualised storage cost 76% lower than that of electricity-based hydrogen. This research also quantifies the critical risk of under-planning by comparing the 22-year NASA data with the 85-year ERA5 dataset (1940-2024) for 14 representative regions. Using the longer dataset increases the required renewable overcapacity by 57% and storage volume by 58%, leading to a cost markup increase of 58–59%. For the cost-optimised case, this means increasing overcapacity from 5.2% to 8.1% and the total balancing system cost markup from 2.9% to 4.5%. To translate these findings on the trade-offs between system cost and infrastructure investment into actionable policy, this work introduces the ‘tau-analogy’, a novel method for identifying balanced solutions. By setting a new standard for data robustness and clarifying key technology choices, this thesis offers a vital framework for designing the physically resilient and economically viable energy systems of the future.