Edge Data Centers as Grid-Serving Infrastructure: A Simulation-Based Assessment of Flexibility Dimensions

Abstract

Edge data centers (DCs) in the 1–5 MW range, connected at the medium-voltage level, have the potential to operate as flexible, grid-serving nodes that couple digital services with electricity grid operation and local heat supply. Prior work has investigated workload shifting, the integration of battery storage and renewables, and waste heat utilization as flexibility options for DC operation. Yet their individual and combined contributions to grid-serving operation have not been sufficiently quantified in a decentralized edge setting. In particular, existing studies have rarely expressed these values in terms that can be directly translated into grid-side instruments such as Flexible Connection Agreements (FCAs). This paper presents a systematic analysis of the potential of these flexibility dimensions, both individually and in combination, on the path toward FCA design. The analysis is based on a Rolling Horizon Optimization that coordinates workload scheduling, renewable integration, battery storage operation, and waste heat utilization under operational constraints. At each decision step, an optimization procedure determines cost-minimizing and peak-load-oriented operating strategies, enabling the systematic assessment of individual and combined flexibility options. The approach is evaluated in a co-simulation framework across a staged set of scenarios, ranging from isolated single-site baselines through individual flexibility dimensions and holistic single-site integration to federated multi-site operation. Performance is assessed using operating cost, peak load, renewable self-consumption, and waste heat utilization as key performance indicators. The results demonstrate how different flexibility dimensions contribute to grid-serving operation and reveal the added value of their coordinated deployment at both single-site and federated scales. These insights provide a foundation for the design of grid-side flexibility instruments, such as FCAs, and support the integration of edge DCs into future renewable-based energy systems.