Abstract:

Unlike traditional flexible loads that have only temporal regulation capability, the loads of multiple data centers can be migrated among them both in time and spatiality, thus becoming a flexible demand response resource with temporal and spatial regulation potential. However, previous studies on load management of multiple data centers have neglected the need for multiple time windows depending on the timeliness in batch load response characteristics. In this regard, we propose a spatio-temporal cooperative operation optimization strategy for multiple data centers considering the load response characteristics. Firstly, the coupling of business and energy flows of multiple data centers is modeled to flexibly configure the number of servers powered on. Then, the temporal and spatial migration characteristics of batch loads are analyzed to establish a multi-data center demand response mechanism. And then, a multi-data center operation optimization model is established with minimum total operation cost as the objective function and user satisfaction as the constraint. Finally, a case study is carried out with multiple data centers as an example. The simulation results show that the proposed scheme can effectively stimulate the demand response potential of multiple data centers, reduce the total operation cost while ensuring user satisfaction, and provide an effective load management strategy for multiple data centers.

Key words: load response characteristics, multiple data centers, spatio-temporal coordination, multiple time windows, business flow and energy flow coupling, demand response