@Article{ee.2025.072631,
AUTHOR = {Luyao Liu, Xiao Liao, Yiqian Li, Shaofeng Zhang},
TITLE = {Multi-Time Scale Optimization Scheduling of Data Center Considering Workload Shift and Refrigeration Regulation},
JOURNAL = {Energy Engineering},
VOLUME = {123},
YEAR = {2026},
NUMBER = {2},
PAGES = {--},
URL = {http://www.techscience.com/energy/v123n2/65668},
ISSN = {1546-0118},
ABSTRACT = {Data center industries have been facing huge energy challenges due to escalating power consumption and associated carbon emissions. In the context of carbon neutrality, the integration of data centers with renewable energy has become a prevailing trend. To advance the renewable energy integration in data centers, it is imperative to thoroughly explore the data centers’ operational flexibility. Computing workloads and refrigeration systems are recognized as two promising flexible resources for power regulation within data center micro-grids. This paper identifies and categorizes delay-tolerant computing workloads into three types (long-running non-interruptible, long-running interruptible, and short-running) and develops mathematical time-shifting models for each. Additionally, this paper examines the thermal dynamics of the computer room and derives a time-varying temperature model coupled to refrigeration power. Building on these models, this paper proposes a two-stage, multi-time scale optimization scheduling framework that jointly coordinates computing workloads time-shift in day-ahead scheduling and refrigeration power control in intra-day dispatch to mitigate renewable variability. A case study demonstrates that the framework effectively enhances the renewable-energy utilization, improves the operational economy of the data center microgrid, and mitigates the impact of renewable power uncertainty. The results highlight the potential of coordinated computing workloads and thermal system flexibility to support greener, more cost-effective data center operation.},
DOI = {10.32604/ee.2025.072631}
}