Distributed Control With Virtual Capacitance for the Voltage Restorations, State of Charge Balancing, and Load Allocations of Heterogeneous Energy Storages in a DC Datacenter Microgrid

• Published in: IEEE transactions on energy conversion Vol. 34; no. 3; pp. 1296 - 1308
• Main Authors: Zhang, Runfan, Hredzak, Branislav, Morstyn, Thomas
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.09.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Allocations; average consensus protocol; Balancing; Batteries; Communication; Computer simulation; Control systems; Controllers; Converters; cooperative control; Data buses; Decentralized control; Distributed generation; Heterogeneous energy storage systems; Impedance; Load sharing; Microgrids; Restoration; State of charge; state of charge balancing; Strategy; Stress concentration; Supercapacitors; virtual capacitive control; virtual impedance control; Voltage control
• DOI: 10.1109/TEC.2018.2889065
• ISSN: 0885-8969

This paper proposes a distributed coordination control strategy for load sharing and energy balancing between heterogeneous energy storages. These control objectives are satisfied through a two-level control structure. At the primary level, the decentralized virtual impedance control, without the requirement of communication links, allocates the low frequency component of the loads to batteries, while the high frequency component is allocated to ultracapacitors. Distributed control strategy, introduced at the secondary level over a sparse communication network, achieves battery state of charge balancing and regulation of the local bus voltages. Two sets of data are exchanged via the communication links, the local bus voltages, and state of charges of batteries. The distributed controller for the restoration of local bus voltages implements an average consensus protocol, while the controller for energy balancing uses a cooperative protocol. In addition, the ultracapacitor voltages are locally restored at a slower time-scale. The proposed control strategy is resilient to communication link failures and features plug-and-play capability. Presented results demonstrate performance of the proposed control strategy for an islanded 380 VDC datacenter with variable loads. Different operating conditions are verified through the RTDS Technologies real-time digital power system simulator using switching converter models and nonlinear battery models.