A Distributed Wireless Testbed for Plug-in Hybrid Electric Vehicle Control Algorithms Ian Beil Ian Hiskens Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor, MI Email: {ianbeil,hiskens}@umich.edu Abstract—There is a desire within the scientific community to use the inherent storage capabilities of plug-in hybrid electric vehicle batteries to enable demand-side management in power systems, increasing the usability of variable generation such as wind and solar power. The potential advantages of increased grid control must be tempered against the harmful effects of charge/discharge cycles on car battery health as well as user constraints on maximum allowable charge time. This paper re- views past research on communication and control architectures for PHEV charging and proposes a novel testbed to allow physical testing of several of the proposed schemes. I. I NTRODUCTION Concerns about climate change due to global warming have spurred research and industry efforts aimed at reducing carbon emissions. To this end, automobile manufacturers have begun selling plug-in hybrid electric vehicles (PHEVs) that use onboard batteries to store energy, resulting in increased fuel economy. By charging their batteries through the electrical grid, PHEVs use less gasoline than conventional vehicles, and may enable the transportation sector to reduce the amount of carbon it produces. A large population of PHEVs could present further benefits to the power system infrastructure through careful utilization of their inherent energy storage capabilities. A well coordi- nated PHEV fleet would allow system operators to perform demand-side management by turning the battery chargers on or off, expanding the controllability of the grid and facilitating an increase in the amount of non-dispatchable generation such as wind and solar that could be accommodated. Battery charging could also provide ancillary services such as frequency reg- ulation and contingency reserves that have traditionally been served by generation units. The potential advantages of increased PHEV penetration can only be realized with the installation of a sufficiently robust, secure, and effective communication and control ar- chitecture. Complicating matters is the reality that the benefits of PHEVs to a grid operator often run counter to those of the vehicle owner, who places constraints on the allowable charge time and seeks to maximize battery health by limiting charge/discharge cycles. This material is based upon work supported by the Department of Energy under Award Number DE-PI0000012, through the Clean Energy Research Center - Clean Vehicle Consortium. This paper introduces a novel testbed designed to assess the performance of PHEV charging schemes using various communication technologies and control architectures. Using hardware that simulates PHEV Li-ion battery loads as well as typical household loads such as refrigerators and other appliances, the testbed will provide a way to assess the efficacy of fully networked charging control strategies. II. RELATED WORK A large research effort has been undertaken towards defining future communication protocols for the electrical grid. Work on communication schemes specifically for PHEV charging is presented in [10], which examines four possible topologies: • HomePlug, a form of broadband over power line com- munication (BB PLC), utilizes the pre-existing grid as a communication platform by placing encoded information on top of the 60 Hz line frequency; • Zigbee, or IEEE 802.15.4, is a mesh network designed for low-power sensor environments; • ZWave is a low data-rate protocol used specifically for home automation applications; • Cellular Networks provide long-range capabilities and are already widely deployed. The research in [10] suggested that cellular encryption schemes are not as robust or secure as those used by the other three topologies. Simulations of one-way load-based charging algorithms found that network traffic on a 1000 car network was well below the throughput capabilities of any of the four proposed communication schemes, although grid-level benefits may not be realized until PHEV penetration reaches higher populations, necessitating larger future simulations. A further study [4] examined another power line scheme: • Narrowband PLC, which enjoys cheaper deployment than BB PLC by avoiding the costs of couplers at the distribution/transmission border, because the narrowband signal can penetrate the transformer (albeit with a noise penalty). Powerline communication schemes face the challenge of needing to filter the harmonics generated by battery charger inverters, but the study concluded that NB PLC could also provide a feasible communication protocol. Any communication scheme will have inherent delays within the network, and the effect of these on a power system 978-1-4673-2308-6/12/$31.00 ©2012 IEEE