综合智慧能源2026，Vol.48Issue(8)：57-68,12.DOI:10.3969/j.issn.2097-0706.2026.08.005
计及差异化需求响应的算电协同优化运行
Optimal operation of computing-electricity coordination considering differentiated demand response
摘要
Abstract
The increasing integration of high-proportion renewable energy has intensified the demand for the regulation of computing-electricity coordination.Meanwhile,the fluctuations in renewable energy output and the growth of data center computing loads have imposed higher requirements for system flexibility,while existing differentiated demand response approaches among multiple stakeholders lack adequate consideration of rolling correction and settlement verification.To address these issues,an optimal operation method for computing-electricity coordination considering differentiated demand response was proposed.The flexible capabilities of differentiated loads,such as battery swapping stations and data centers,were classified into three types of demand response services:local load curtailment,temporal shifting,and spatial migration.These services enabled load reduction during events,deferrable task transfer,and cross-regional load redistribution,respectively.Additionally,a market-oriented interface model was constructed.A nodal response price model was established,with factors such as voltage risk,branch congestion,and peak-load margin incorporated into price signals to verify the physical feasibility of demand response clearing results.Nodal response quantities were also verified using power-flow and security constraints to prevent voltage limit violations and branch overloading.On this basis,a distributed clearing method based on the alternating direction method of multipliers was designed.While preserving local constraints on inventory,tasks,service-level agreements,state of charge,and quality of service,the method achieved coordination among multiple stakeholders through the exchange of common variables.A rolling price correction mechanism was proposed based on measured responses,residual states,and recovery risks.Case study results under different strategies and in both medium-voltage and medium-to-high-voltage distribution network scenarios showed that the proposed method effectively reduced response shortfalls,mitigated voltage limit violations,reduced network losses,improved demand response fulfillment quality,and enhanced safety,economic performance,and settlement stability during dynamic execution.The method transformed differentiated flexible resources into market-oriented response services that can be procured,delivered,settled,and verified.The medium-voltage case validated the effectiveness of different combinations of demand response services and algorithmic stages,whereas the medium-to-high-voltage case examined the engineering applicability of the method under multiple voltage levels and cross-zone migration conditions.The findings provide support for the coordinated operation of computing loads and distribution networks under high-proportion renewable energy conditions.关键词
数据中心/算电协同/需求响应/差异化建模/交替方向乘子法Key words
data center/computing-electricity coordination/demand response/differentiated modeling/alternating direction method of multipliers分类
能源科技引用本文复制引用
张冬冬,刘一宁,陈拓宇,武晓栋,江美慧,朱虹谕,张杰..计及差异化需求响应的算电协同优化运行[J].综合智慧能源,2026,48(8):57-68,12.基金项目
国家自然科学基金项目(52677064) （52677064）
国家重点研发计划项目(2026YFE0211200) （2026YFE0211200）
内蒙古自治区自然科学基金项目(2026QC0364,2026QC0321)National Natural Science Foundation of China(52677064) （2026QC0364,2026QC0321）
National Key R&D Program Project(2026YFE0211200) （2026YFE0211200）
Inner Mongolia Natural Science Foundation(2026QC0364,2026QC0321) （2026QC0364,2026QC0321）
