Jiangjiang Wang

The strong coupling among computation, electricity, and cooling in data centers poses challenges for operational optimization. Meanwhile, uninterruptible power supply systems remain underutilized, with their energy storage potential not fully exploited. To address these issues, this paper proposes a Stackelberg game framework incorporating shared energy storage based on an energy storage-based uninterruptible power supply for coordinated energy–computing management when distributed energy systems interact with data centers. The distributed energy operator acts as the leader to determine pricing and device dispatch, while the data center, as the follower, minimizes operating cost through workload scheduling. The energy storage-based uninterruptible power supply serves as shared energy storage to facilitate interaction between the two parties. Comparative case studies with and without game interaction and energy storage-based uninterruptible power supply participation are conducted to analyze impacts on pricing strategies, economic performance, and coordination between energy supply and computational demand. The results show that, compared with conventional grid-based supply, the proposed framework reduces data center operating cost by 16.72%. With energy storage-based uninterruptible power supply integration, the cost is further reduced by 1.60%, while distributed energy operator profit increases by 3.93% and photovoltaic curtailment penalty decreases by 57.69%. The proposed model provides support for decision-making between data centers and distributed energy systems.