With carbon neutrality as a target and the increased penetration of renewable energy, the operational flexibility of power systems has begun to face challenges. In order to explicitly represent the operational flexibility of power systems, two types of flexibility indexes and corresponding models for their evaluation are established in this paper. One of the indexes is the supply–demand balance, which evaluates the adequacy of operational flexibility at the system level. The other is the availability of flexible resources, which comprehensively quantifies the flexibility of the power system from the perspectives of power generation, load, and energy storage. In the case study presented here, the proposed evaluation method is illustrated and validated based on a provincial power system in China. Next, the role of energy storage in enhancing flexibility is quantitatively analyzed using the proposed indexes. Then, the economic model reveals the nonlinear decline in the marginal benefit of investment in energy storage. Energy storage alone cannot fully meet the requirements for supply–demand balance in the power system, necessitating a comprehensive consideration of the available capacity for flexibility from the perspectives of generation, load, and energy storage. Analysis of a typical scenario shows that the provincial power system has 5000 MW of upward and downward flexibility in capacity. The numerical results highlight the critical importance of integrating flexibility across all components.
The transition toward sustainable power systems with high shares of renewable energy requires flexible resources that can support both resource adequacy and broader sustainability goals. Battery energy storage is an important enabler of this transition, but capacity-linked subsidies may alter its operation when payments require peak-period availability. Reserving energy for this obligation sacrifices spot market arbitrage value, and so the subsidy needed to induce capacity commitment depends on an endogenous availability opportunity cost. This study aims to characterize that trade-off and derive a practical rule for subsidy calibration. We formulate an annual unified capacity commitment model in which storage selects one committed capacity across seasonal representative days while optimizing hourly dispatch subject to peak-period energy reservation constraints. The model is reduced to a one-dimensional problem that balances linear subsidy revenue against a convex, piecewise affine opportunity cost function, yielding forward commitment thresholds and an inverse rule from a target commitment to the minimum subsidy required to induce it. In a representative system calibrated for Gansu, commitment remains zero below , reaches 46.3% of maximum capacity at , and becomes full at . The average opportunity cost threshold is . A target of requires but implements ; at high subsidies, normalized throughput falls from 2.001 to 1.298, and the spot revenue share declines from 1.000 to 0.265. These results show that effective capacity subsidy design must account for opportunity cost thresholds, staged commitment, over-procurement, and reduced spot market participation.