Abstract
<title>Abstract</title> <p>Battery energy storage systems (BESS) are increasingly recognised as an enabling technology for renewable-rich electricity systems; however, their sustainability value is often assessed using simplified or externally defined price signals rather than the actual economic structure of the power system. This study develops a KKT-consistent mixed-integer optimisation framework that embeds battery storage directly into a national electricity dispatch model, allowing marginal electricity prices, generator dispatch, and storage operation to be determined simultaneously. Building on the marginal-cost pricing principle established by Schweppe et al., the proposed direct-embedding approach reduces computational complexity compared with conventional big-M complementarity reformulations while preserving the optimality conditions of the underlying bi-level formulation. The framework is applied to Sri Lanka’s national electricity system using a corrected 18-plant generator fleet calibrated with real 15-minute dispatch data, together with government-reported solar, wind, and curtailment records. A 240 MW battery system is evaluated under single-day, multi-day, and sizing scenarios. Results demonstrate that storage responds economically to system conditions by charging during periods of high renewable availability and discharging during higher marginal-cost periods, generating 7.21 million LKR in arbitrage revenue across four validated reference days and 42.7 million LKR over a 15-day continuous assessment period. Battery integration also reduces renewable energy wastage by recovering a portion of observed curtailment and provides an estimate of avoided thermal generation externalities. However, comparison with reported curtailment volumes indicates that a single 240 MW battery captures only a limited share of renewable oversupply, highlighting the importance of storage planning based on actual system flexibility requirements rather than average operating conditions. The proposed framework provides a data-driven approach for evaluating the role of battery storage in supporting sustainable renewable integration in transitioning electricity systems.</p>