Our paper on online EV scheduling for adaptive charging networks [1] has been accepted to IEEE Transactions on Sustainable Computing!
![[doi]](https://mallada.ece.jhu.edu/wp-content/plugins/papercite/img/external.png){kind=small}
A. Bahram, M. H. Hajiesmaili, Z. Lee, N. Crespi, and E. Mallada, “Online EV Scheduling Algorithms for Adaptive Charging Networks with Global Peak Constraints,” IEEE Transactions on Sustainable Computing, 2020. [Bibtex] [Abstract] [Download PDF]
Electricity bill constitutes a significant portion of operational costs for large scale data centers. Empowering data centers with on-site storages can reduce the electricity bill by shaping the energy procurement from deregulated electricity markets with real-time price fluctuations. This paper focuses on designing energy procurement and storage management strategies to minimize the electricity bill of storage-assisted data centers. Designing such strategies is challenging since the net energy demand of the data center and electricity market prices are not known in advance, and the underlying problem is coupled over time due to evolution of the storage level. Using competitive ratio as the performance measure, we propose an online algorithm that determines the energy procurement and storage management strategies using a threshold based policy. Our algorithm achieves the optimal competitive ratio of as a function of the price fluctuation ratio. We validate the algorithm using data traces from electricity markets and data-center energy demands. The results show that our algorithm achieves close to the offline optimal performance and outperforms existing alternatives.
@article{bhlcm2019tsusc,
abstract = {Electricity bill constitutes a significant portion of operational costs for large scale data centers. Empowering data centers with on-site storages can reduce the electricity bill by shaping the energy procurement from deregulated electricity markets with real-time price fluctuations. This paper focuses on designing energy procurement and storage management strategies to minimize the electricity bill of storage-assisted data centers. Designing such strategies is challenging since the net energy demand of the data center and electricity market prices are not known in advance, and the underlying problem is coupled over time due to evolution of the storage level. Using competitive ratio as the performance measure, we propose an online algorithm that determines the energy procurement and storage management strategies using a threshold based policy. Our algorithm achieves the optimal competitive ratio of as a function of the price fluctuation ratio. We validate the algorithm using data traces from electricity markets and data-center energy demands. The results show that our algorithm achieves close to the offline optimal performance and outperforms existing alternatives.},
author = {Bahram, Alina and Hajiesmaili, Mohammad H. and Lee, Zachary and Crespi, Noel and Mallada, Enrique},
doi = {10.1109/TSUSC.2020.2979854},
grants = {CAREER-1752362, CPS-1544771, ENERGISE-DE-EE0008006, AMPS-1736448, TRIPODS-1934979,EPCN-1711188,},
journal = {IEEE Transactions on Sustainable Computing},
month = {1},
title = {Online EV Scheduling Algorithms for Adaptive Charging Networks with Global Peak Constraints},
url = {https://mallada.ece.jhu.edu/pubs/2020-TSUSC-BHLCM.pdf},
year = {2020}
}