With the development of information technology, the scale of data center infrastructure has become larger and larger, which has always brought many challenges, such as unbalanced power load, high energy consumption, small scalable space and so on. In recent years, a distributed power supply structure, namely microbus power supply, has emerged, which is different from the traditional power distribution mode in computer room. It can adjust the power load distribution online in real time. This paper analyses the advantages of this micro bus structure power supply mode, and optimizes the cabinet power collection strategy based on branch and bound algorithm. By modeling and simulating the power consumption of the computer room in a real data center, the experimental results show that the proposed algorithm can greatly reduce the unbalance rate of the three-phase power supply equipment in the upper end of the computer room. Compared with the traditional power supply mode, it can reduce the maximum power imbalance by 49.88%.
[1]
Jun Tian,et al.
Dynamic Programming for NP-Hard Problems
,
2011
.
[2]
Jeffrey E. Schaller,et al.
Branch-and-bound algorithms for minimizing total earliness and tardiness in a two-machine permutation flow shop with unforced idle allowed
,
2019,
Comput. Oper. Res..
[3]
Vikrant Vaze,et al.
An integer programming approach to fisheries observer deployment
,
2019,
Transportation Research Part E: Logistics and Transportation Review.
[4]
Sunghyok Kim,et al.
Distributed virtual negative-sequence impedance control for accurate imbalance power sharing in islanded microgrids
,
2018
.
[5]
Juan Carlos Burgos,et al.
Rules to estimate the expected values of zero-sequence impedances in 3-phase core-type transformers
,
2018,
Electric Power Systems Research.