A prediction and load shed-based approach of controlling a medium voltage AC/DC testbed

Abstract

Shipboard power systems have previously relied on oversized generators with minimal monitoring of the output, preventing the implementation of many real-time controls. Redundant generation is implemented instead of controls, resulting in uncontrolled load sheds when capacity is exceeded, or generators fail. The naval power system community is extensively studying the advantages of distributed zonal power system architectures that allow for the buffering of multiple generation sources and loads using energy storage and regulating power electronics. Extensive sensing and controls are necessary for successful operation of a distributed power system, and they must be evaluated adjoint to representative hardware topologies. The University of Texas at Arlington (UTA) is utilising a medium voltage (MV) AC/DC testbed to examine predictive high ramp rate (PHRR) and advanced load shed (ALS) control algorithms developed and modelled by Clarkson University (CU). A single zone of a zonal power system is being emulated using UTA’s testbed. Experimental results collected after integrating the CU control strategies into the UTA testbed will be discussed here.