Impacts of the scenario temporal structure and data variability on the planning and reliability of integrated energy system coupling electricity–thermal–hydrogen

Reliable and efficient integrated energy system (IES) design depends critically on how operating scenarios are represented, as scenario assumptions directly govern projected performance and robustness. Beyond uncertainty itself, temporal continuity and data variability are key factors shaping planning outcomes. This study proposes a scenario-driven optimization framework that combines multi-objective operational scheduling with an adaptive regional search-and-contraction algorithm for high-dimensional IES planning. Eight representative scenario sets with different temporal structures and variability levels are constructed and applied to a multi-energy hub integrating electricity, heat, hydrogen, and gas. Results show that short-duration scenarios systematically distort planning outcomes by overestimating renewable capacity, underestimating storage requirements and ageing-related costs, and thus yielding overly optimistic reliability evaluations. In contrast, medium-duration scenarios better capture seasonal and inter-day dynamics, leading to more balanced capacity allocations and lower risk-adjusted costs. Further extension of scenario length offers only marginal improvements in overall planning performance, indicating a practical trade-off between computational burden and planning accuracy. An indicator system is finally developed to quantify scenario characteristics, and correlation analysis further clarifies how statistical properties, resource-load matching, and extreme events jointly influence capacity allocation and total cost.