Cycle-aware optimization of battery storage in distribution networks under load and market uncertainty

Abstract

The accelerating integration of renewable-based distributed generation into distribution networks is reshaping their operational landscape, introducing variability, forecast uncertainty, and limited dispatchability that collectively strain stability, reliability, and economic performance. Battery Energy Storage Systems (BESS) provide a versatile means to address these challenges. Their long-term effectiveness, however, hinges on accurate degradation modeling, appropriately scaled capacity, and scheduling strategies that remain robust under uncertain operating conditions. This study develops a multi-stage Mixed-Integer Linear Programming (MILP) framework that jointly optimizes operational cost and asset longevity by embedding a cycle-aware degradation model into a scenario-based stochastic scheduling structure. Variability in demand, photovoltaic generation, and electricity market prices is represented through probabilistic scenario generation and reduction, ensuring operational robustness to market and resource fluctuations. The proposed optimization framework was initially validated on the IEEE-33 bus test feeder using high-resolution demand profiles obtained from a regional distribution system operator, hourly photovoltaic generation data sourced from utility-scale plants located within the same geographical area, and officially verified hourly market price series retrieved from the Transparency Platform of the Turkish Energy Exchange (EPI(center dot)AS,). Subsequently, employing the identical dataset, the framework was applied to a real medium-voltage distribution feeder comprising 61 buses, wherein extensive sensitivity analyses were conducted under diverse operating conditions and parameter variations. The approach also incorporates a post-simulation technical loss assessment to validate the physical feasibility of the optimized strategies. Overall, the framework offers a rigorous, scalable, and practically applicable decision-support tool for integrating storage into renewable-rich distribution systems under realistic economic and technical conditions.