A Study on Data Center Two - layer Optimal Scheduling Based on Carbon - electricity Synergy
Qingsheng Li, Junqiu Fan, Xueyong Tang, Weixing Zhao, Long Yuan, Ning Nan, Qingming Zhao, Wenyong Tang · 2025
The problem of high energy consumption and carbon emissions of rapidly growing data centers is widely concerned. Typically, carbon emissions in data centers are calculated by multiplying the electricity consumption by a fixed carbon emission factor. However, this approach ignores the varying proportions of renewable energy in the power grid at different times, making it challenging to fully inspire data centers to optimize their energy using. Accordingly, this paper first constructs a computing power-carbon flow analysis model to realize the dynamic evaluation of green power consumption in data centers. On this basis, this paper further proposes a bi-level optimal scheduling model for carbon and electricity between interconnected data centers and power systems. The upper model is for power system operator to minimize the total cost of grid operation, which determines the power as well as carbon emission flow in power system. The lower-level model focuses on carbon-electricity collaborative demand response within interconnected data centers, aiming to minimize their comprehensive electricity carbon cost. It optimizes the computing load processing plans and operation of distributed generator as well as energy storage equipment in data centers. Upon receiving the value signal from the distribution system operator, the data operator considers its own carbon and electricity characteristics to achieve rational energy using. Finally, the results obtained from the improved IEEE-14 system demonstrate that the proposed model effectively reduces energy consumption and carbon emission of data centers.