Two-stage robust optimal configuration of a data center integrated energy system with waste heat recovery and carbon capture

The rapid expansion of the digital economy has intensified the challenges of high energy consumption and carbon emissions associated with data centers, making the development of data center integrated energy systems (DCIES) a critical pathway toward a low-carbon transition. However, achieving deep decarbonization in DCIES while ensuring the efficient and economic utilization of data center waste heat remains a significant challenge. To address these issues, this study proposes a two-stage robust optimal configuration framework for a data center–carbon capture integrated energy system (DC-CC-IES). First, a novel waste heat recovery and utilization architecture is developed, in which a high-temperature heat pump upgrades low-grade waste heat to supply the regeneration process of carbon capture, thereby enabling waste-heat-driven decarbonization. Second, a unified integrated demand response mechanism is established to coordinate computing workloads with electricity, heating, and cooling loads, significantly enhancing system flexibility. Finally, to address uncertainties in renewable generation, multi-energy demands, and data center workloads, a two-stage robust optimization model is formulated based on a polyhedral uncertainty set constructed via Latin hypercube sampling. Case studies demonstrate that, compared with the baseline scenario, the proposed approach reduces total costs by 14.02%, decreases carbon emissions by 43.38%, improves energy efficiency by 7.4%, and increases renewable energy utilization by 15.8%, while ensuring 100% energy supply reliability. Overall, this work provides a novel system architecture and a robust planning methodology to support the low-carbon and sustainable development of DCIES under multiple uncertainties.