Cross-vector flexibility in sector-coupled low-carbon energy systems

This paper holistically assesses the current literature by examining how cross-vector flexibility is observed, coordinated, and verified in low-carbon energy systems based on sector coupling. To this end, the interplay between cross-vector flexibility, digital twins and artificial intelligence (AI)-enabled observability, predictive control approaches, local coordination mechanisms, and flexibility markets is addressed within a common analytical framework. This review shows that cross-vector flexibility is more than an energy conversion technology problem; it is also a coordination problem. Following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) framework, a systematic Scopus search and screening process yielded an analytical core of 107 studies. Within this core, only 19 studies (17.8%) progress beyond synthetic simulation: thirteen use real or case-based data, six provide co-simulation, hardware-in-the-loop or closed-loop field evidence, and none verifies delivered flexibility against a baseline. Observability is addressed as a primary concern in only seven of the 107 studies, making it the least represented layer of the coordination stack. In addition, 88 studies (82.2% of the core) represent no socio-technical conditions. The value of flexibility depends not only on the capacity to shift, store, or convert energy, but also on the ability to monitor system states, ensure coordination between different actors and energy carriers, verify the service provided, and create mechanisms by which flexibility can be procured and evaluated as a service. When these layers are considered together, the observability and validation deficiencies spread upwards throughout the coordination stack, which limits the applicability of advanced control approaches and market designs in real systems. Therefore, greater emphasis should be placed on observability infrastructures that support validation, transparent interface definitions between coordination layers, validation processes that extend beyond simulation, and assessment frameworks that explicitly consider behavioral and institutional realities. Developing these elements is critical to transforming cross-vector flexibility from a concept primarily assessed in simulation environments into a verifiable, transferable, and operationally deployable service in decarbonized multi-energy systems.