面向异构数据中心的低碳负荷准线生成与分解方法
By constructing a nonlinear energy consumption model and a virtual power node model for heterogeneous data centers, a low-carbon load baseline is generated and decomposed, solving the problems of accuracy and low-carbon scheduling in data center load regulation, and realizing precise mapping and optimized allocation of load regulation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIV
- Filing Date
- 2026-04-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing data center demand response technologies lack a continuous and stable load baseline, cannot accurately reflect the intrinsic relationship between computing power tasks and power consumption, ignore the multi-factor coupling effect of server utilization and cooling systems, and do not fully consider the differences between heterogeneous data centers, resulting in deviations in capacity estimation and mismatches in control instructions, making it difficult to achieve accurate mapping and optimized allocation of low-carbon scheduling.
A nonlinear energy consumption model for heterogeneous data centers is constructed to determine the target energy efficiency ratio of IT equipment and cooling systems. A virtual power node model is established, and a low-carbon load baseline is generated based on a cost minimization model. The baseline is then decomposed into data centers through a baseline decomposition optimization model, and low-carbon load adjustment is achieved by combining computing power task migration instructions.
It achieves precise mapping and optimized allocation of system-level low-carbon load baselines to each data center, effectively guiding data center loads to participate in low-carbon scheduling and improving the execution accuracy and reliability of demand response.
Smart Images
Figure CN122051949B_ABST