1

The Impact of Plug-in Hybrid Electric Vehicle Interaction with Energy Storage and Solar Panels on the Grid for a Zero Energy House
A. Bedir, Student Member, IEEE, B. Ozpineci, Senior Member, IEEE, and J. E. Christian

Abstract- Renewable energy sources and plug-in hybrid electric vehicles (PHEVs) are becoming very popular in research areas, as well as in the market. The aim of this paper is to demonstrate how a solar powered building interacts with energy storage and how it can be used to power a PHEV and to support the grid with peak shaving, load shifting, and reducing annual energy usage. A net zero energy house (ZEH5) is selected as the base house for this experiment. Oak Ridge National Laboratory (ORNL) is developing simulation models and energy management scenarios using the actual solar production and residential energy usage data, and a PHEV. The system interaction with the grid is evaluated after getting all the data from PHEV charging, photovoltaic (PV) power production, and residential load. Index TermsRenewable integration, PHEV, peak shaving, smart charging, smart grid, zero energy house, energy management.

I. INTRODUCTION The renewable energy sources and plug-in hybrid electric vehicles (PHEVs) are becoming very popular because of the combination of high fuel costs and concerns about emission issues. Also, the dependence on fossil fuels threatens the environmental sustainability and energy security for all around the world. However, the building and transportation areas demand around 70% of the United States energy [1]. A building and transportation structure that uses a renewable energy system approach with sustainable planning, zero energy houses, advanced vehicles, and smart grid applications could significantly decrease this energy usage, carbon emissions, and climate change impacts worldwide. Oak Ridge National Laboratory (ORNL) is developing simulation models and energy management scenarios for a net zero energy house using the actual solar power production, residential energy usage data and an actual PHEV.
This work was supported in part by Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U.S. Department of Energy under Contract DE-AC05-00OR22725, and in part by the U.S. Government under Contract DE-AC05-00OR22725. A. Bedir is with the Department of Electrical Engineering, Tennessee Technological University, Cookeville, TN 38501 USA (e-mail: abedir42@tntech.edu) B. Ozpineci and J. E. Christian are with Oak Ridge National Laboratory, Oak Ridge, TN 37831 USA (e-mail: ozpinecib@ornl.gov, christianje@ornl.gov)

Previous studies on this topic consist of the potential impacts of PHEVs on regional power generation and on utility systems in large-scale demonstrations [2, 3]. However, this study will show how a solar powered building interacts with energy storage and how it can be used to charge the PHEV battery, to support the grid with peak shaving and load shifting, and to reduce annual energy usage. It is important to define PHEV and zero-energy house before analyzing the system. A PHEV, like an HEV, has an electric motor and a gasoline engine, but it has a larger battery than a regular hybrid electric vehicle and can go on all-electric mode for a specified distance. The PHEV battery can be recharged by plugging it into a conventional electrical outlet [4]. As the numbers of PHEVs increase, their impact on the grid will be more pronounced. If drivers control the timing of charging, and if they plug in when convenient rather than when utilities prefer they charge their vehicles, PHEVs could cause local or regional constraints on the grid and require additional capacity [3]. The recent studies suggest that by 2030 almost all regions of United States (10 out of 13) will need to add capacity for charging PHEVs, mostly in the situations where PHEVs are charged at 6 kW during the evenings [3]. To avoid these problems the utilities would need to expand their capacity, increase their imports, or establish demand response programs. However, smart-grid applications with smart charging can be a key solution to grid overloading problems.

Fig 1. Illustration of the designed system

978-1-4244-6547-7/10/$26.00 2010 IEEE

Zero-energy houses (ZEHs) are residential buildings that can generate as much energy on site as they consume. The ZEH goal is being pursued at ORNL by combining energy efficient technologies and methods, on-site photovoltaic (PV) power generation, and smart energy-management appliances. Fig. 2 shows a comparison of monthly average loads of a ZEH (in green) with typical house loading (in blue). Power consumption of the ZEH is within a range of 750 2000W where a typical house has energy power consumption between 2250 6250W.

ZEH5 has twelve 208W (2.5kWp) Sharp PV panels installed on the south-facing roof at a 4/12 pitch (~18 degrees) covering 210 square feet [7]. The DC voltage coming from the PV is terminated to a Gridpoint Connect (Model C36-10-G2) appliance, which has an outback inverter, for grid interface, and a 10 kWh of leadacid battery for energy storage (ESS) [8]. This appliance uses the PV power to charge the batteries, supply energy to the house, or feed excess energy to the grid. The stored energy in the batteries can be dispatched to the grid during peak load times or used to supply energy to the critical circuits during power outages.

Fig. 4. Monthly PV generation and whole-house consumption for ZEH5 between Jun 24, 2008, and Feb 23, 2008 [6]. Fig. 2. Loading comparison of a ZEH with a typical one.

II. SYSTEM SETUP For the PHEV and ZEH system interaction experiment, ORNLs fifth zero-energy house in Lenoir city, Tennessee, code named ZEH5 was selected as the test base.

A Toyota Prius plug-in conversion vehicle that has a 5kWh lithium-ion Hymotion battery is used in the study as the PHEV to demonstrate the impact of PHEVs. The battery pack in this PHEV has 616 cells with 56 in series and 11 in parallel [9]. The charging characteristics of this Prius PHEV are plotted in Fig. 5 after an 11-mile drive. The PHEV battery draws around 1200W for 2.25 hrs.

Fig. 3. System Structure

Fig. 3 illustrates the overall energy management system using the GridPoint platform that interfaces the PV panels, energy storage technologies, and renewable energy sources with the electric grid [5]. The energy efficient building technologies used in ZEH5 coupled with the PV system can reduce summer peaks by 40% and supply about 29% of the houses energy needs for the mixed-humid climate. Without any demand-side management features such as controls, battery storage, or pre-cooling, this near-zero-energy home can save about 40% energy compared to the same size house with identical occupancy usage patterns. About half of these savings come from energy efficiency features and the other half from the PV [6]. The power production of the solar panels for ZEH5 is plotted in Fig 4.

Fig. 5. Measured PHEV battery charging characteristics for the 5-kWh Hymotion battery

After acquiring the PHEV and completing the set-up of the testing module, a regular driving schedule is followed for daily charging of the vehicle at ZEH5 and on the ORNL campus during the summer peak energy demand periods. The Gridpoint device is set at ZEH5 to capture the residential loading data as well as the PV generation data. This data also include the PHEV battery and the GridPoint internal battery (ESS) charging data at night. A second PHEV charging station is located at the ORNL main campus where the PHEV is charged on ORNL campus after it is driven 11 miles from the ZEH5 to ORNL. Gasoline use by the vehicle is also tracked in order to evaluate the full performance and efficiency of the vehicle and for electricity and gasoline usage comparison. III. PERFORMANCE EVALUATION For a residential building with solar panels and energy storage, the impact of PHEV charging on the grid would depend on the capacity of the PHEV battery, PV, and residential ESS. The ESS in the building can be used to partially or fully charge the PHEV battery to reduce the peak load due to PHEV battery charging. The integration of the PHEV, ESS, and PV with residential load of a family of five is evaluated by using real system (residential load, PV, and ESS charging) data and an energy management model developed in Matlab-Simulink. For the simulation, the PHEV charging characteristics shown in Fig. 5 are used to show the impact of the PHEV. The model has been simulated for four different cases to show the impact of each energy component: 1residential load and PHEV, 2- residential load, PHEV, and PV, 3- residential load, PHEV, and ESS, 4- residential load, PHEV, PV, and ESS. The next three subsections will show the power demand case results for the peak day in June 2009 (turns out to be the 12th), the daily average for the month of the same month, and the each single day of the month of June. A. Peak day (June 12, 2009) The peak day loading characteristic for June 2009 is selected as an initial step of the ZEH5 system analysis. The peak power demand days are extremely important for utilities to determine the capacity of the grid, and electricity demand of the customers. Any extra instantaneous loading occurring during the peak days can cause some overloading situations or contingencies on the grid. In Fig. 6, the red lines show the residential load on June 12, 2009, and the blue lines show the new load characteristics with the new component added. Case 1: The PHEV is plugged in at 7pm after peak hours (5-7pm) for TVA [6]. The peak power demand without the PHEV is around 9200 Watts and after PHEV starts to charge, the peak load increases to around 10400 Watts, an increase of 1200W. Case 2: In this case, PV is added to the system and is producing power from 7am to 6pm. This plot shows that the PV can feed up to 1000W power to the grid during this time but it cannot offset the peaks generated by the PHEV charging since there is no PV power generation during PHEV charging.

Case 3: This case adds a 10 kWh battery ESS to case 1 with no PV. The ESS is charged with 1000W during the offpeak hours, 1am to 11am, and discharged with 2.5kW at the on-peak time and during the PHEV charging time, 5pm to 9pm; therefore, it can shave the peaks during peak hours. Since the PHEV battery needs around 1200W to charge, the ESS can solely be used to charge the PHEV battery without an extra power from the grid. This would not be the case for ever PHEV because it depends on the capacity of the PHEV battery.

Fig. 6. ZEH5 loading for June 12, the peak day in June 2009.

Case 4: This case integrates both the 2.5 kWp PV and 10 kWh ESS with the house. With all these components, ZEH5 can sell energy back, reduce the peak power demand, and offset the PHEV charging power needs. B. Daily average over June 2009 The daily average power demand profile is a very important factor in understanding the energy-consumption characteristics of a house. It can be observed from Fig. 7 that the average peak loading is low compared to the typical peak days for the monthly average data for this case. The peak power demand for an average day in June 2009 is around 3.7 kW compared to the highest daily peak 9-10kW on June 12, 2009. Case 1: On average, the PHEV charging adds 900W to the load power increasing the demand from an average of 3700 W to 4600 W. This means that PHEV charging increases the

houses peak load on average by about 27%. Case 2: On average the PV sends up to 1000W to the grid. Case 3: The ESS has a significant effect on the monthly average load. It shaves the peaks caused by the PHEV and residential load during peak hours up to 2.5kW. Case 4: With all system components added on, the ZEH5 can sell energy back, reduce the peak power demand, and offset the PHEV charging power needs.

Case 4: This system has both the advantages of case 2 and case 3. There is solar power contribution, peak shaving, and PHEV load offsetting by the integration of PV and ESS.

Fig. 7. ZEH5 average loading for June 2009.

C. 30 days worth of data for June 2009 Fig. 8 shows all the data, 24 hours per day for 30 days, for June 2009 in one plot where the x-axis shows the day of the month and y-axis shows the hour of the day. Case 1: The residential load and PHEV charging for the month of June are shown together. Red contours show the peak loading between 8kW to 10 kW. It can be seen that the 10th and 15th days of the June 2009 were the peak days and there are obvious peaks occurring due to PHEV charging around 8pm Case 2: The impact of PV on the power demand is shown in this plot. PHEV charging at 7pm is still significant since the solar power production is not available after 6pm. The solar power production is available between 10 am and 3 pm shown by dark blue color as the power sent to the grid. The most energy efficient days were the days between June 6th and 10th because the power demand peaks are lower with less red and more darker blues showing they were sunny days. Case 3: The ESS has a significant effect on the peak reduction. Not many peaks can be observed even in the evening hours after 7pm when the PHEV is being charged.

Fig. 8. ZEH5 Loading for all days of the June09.

IV. EXPERIMENTAL RESULTS The previous sections showed the interaction of the residential power demand, PV, ESS, and the PHEV using a simulation model showing different possible cases. The simulation model had the actual data for PV generation, ESS storage, and residential load but did not have actual daily charging data of the PHEV, it used one set of data that was obtained separately. This section has experimental results obtained from Gridpoint data feed system with the actual PHEV connected to the system. The PHEV was set to start charging at 8 pm on July 22, 2009. The ESS is not used to show the effect of PHEV charging on a typical summer load for a near zero energy house. The ESS of ZEH5 is being used for only peak days to keep the peak of the house at a certain level. Future test results will also be used to verify the results from earlier sections of this paper. Fig 9 shows that the peak of the house load can increase from 3.2 kW to 4 kW with the PHEV charging.

PV generates power during the daytime and it can reduce the load by around 2kW but they cannot offset the PHEV load.

Fig. 9. Experimental results for ZEH5s one day loading on July 22, 2009.

V. PRICE EVALUATION For the price evaluation, a time based rate schedule has been used that should enable the utility customer to manage energy use and cost through advanced metering and communication technology [10]. A real-time pricing schedule has been derived from California Alternate Rates for Energy (CARE) pricing program of the Pacific Gas and Electric (PG&E) company [11]. TVAs Green Power Switch (GPS) program ratings have also been considered [12]. Fig. 10 shows the derived real-time pricing schedule and TVAs fixed pricing rate. The real-time pricing varies from 7 to 16 cents/kW (blue) and its average has been fixed to 10 cents/kW (red).

Fig. 11 shows the daily energy cost for ZEH5 for fixed rate pricing (FRP) and estimated real-time pricing (RTP), for the cases studied earlier: No PHEV case: only residential load, Case1: residential load and PHEV, Case2: residential load, PV and PHEV, Case3: residential load, ESS, and PHEV (ESS charged during the off-peak hours and discharged during the peak hours), Case4: house load, ESS, PV, and PHEV. As seen in the figure, the energy cost is higher for real-time pricing than fixed pricing because the peak loading of the house including some of the PHEV charging time occurs in parallel with the utility peak. Note that this study assumes no additional incentives for solar production and sell back. With this assumption, only in cases 3 and 4 is the situation slightly different, because of the peak shaving due to the ESS in these cases. VI. CONCLUSION A solar-powered building interacting with an energy storage system and a PHEV battery is evaluated using real data from a zero-energy house and an actual PHEV with 5kWh battery capacity. The system is compared for four different cases for providing grid support with peak shaving, load shifting, and reducing annual energy usage. Experimental results showed that the PHEV can increase the peak power demand of the house which can be offset using an ESS to charge the PHEV battery. With real-time pricing, the ESS also helps reduce the cost of daily energy use. If the solar power sent to the grid were priced differently, the PV power generation during the day would have further reduced the cost. VII. REFERENCES
[1] N. Carlisle, J. Elling, and T. Penney, A renewable energy community: key elements, NREL, Boulder, CO, Tech Rep.TP-540-42774, Jan 2008. [2] P. Kadurek, C. Ioakimidis and P. Ferrao, Electric vehicles and their impact to the electric grid in isolated system, in IEEE Int. POWERENG Conf., pp. 49-54. [3] S. W. Hadley, A. Tsvetkova, Potential impacts of plug-in hybrid electric vehicles on regional power generation, ORNL, Oak Ridge, TN, Tech Rep., TM-2007/150, Jan 2008. [4] Rousseau, A. Pagerit, S., Gao, D., "Plug-in hybrid electric vehicle control strategy parameter optimization," in 23rd International Electric Vehicle Symp. (EVS23), Anaheim, CA, Dec. 2007. [5] Gridpoint Smart Grid Solutions, (2009, Aug.). Gridpoint products web site. [Online]. Available: http://gridpoint.com/ [6] J. Christian, Zero energy peak from zero energy homes, in ACEEE Summer Study on Energy Efficiency in Buildings, pp. 1.37-1.48. [7] Sharp Solar Electricity, (2009, Aug.). [Online]. Available: http://solar.sharpusa.com/solar/home/1,2462,,00.html. [8] P. Boudreaux, T. Gehl, and J. Christian, The role of solar energy toward the goal of zero-energy homes in USA mixed-humid climates, in Solar 2009, Buffalo, NY, May 2009. [9] Hymotion-A123 Battery Company, (2009, Aug.). [Online]. Available: http://www.a123systems.com/hymotion/aboutus. [10] TVA Staff, Time-based metering and communications, Tennessee Valley Authority - Energy Policy Action, Knoxville, TN Tech. Rep. 2005. [11] A. Moholkar, P. Klinkhachorn and A. Feliachi, Effects of dynamic pricing on residential electricity bill, in IEEE Power Systems Con. and Exp., pp. 1030-1035. [12] Tennessee Valley Authority, (2009, Aug.) Green Power Switch [Online]. Available: http://www.tva.gov/greenpowerswitch/.

Fig. 10. Pricing schedules.

5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0

Price in Dollars ($)

No PHEV 3.74 4.33

Case 1 4 4.55

Case 2 2.47 2.79

Case3 4 3.51

Case4 2.54 1.8

FPP RTP

Fig. 11. Real-time and fixed pricing comparison for daily average load of June 2009.

VIII. BIOGRAPHIES
Abdulkadir Bedir (S'08) joined in Tennessee Tech University (TTU) as a graduate student in Electrical and Computer Engineering Department in August 2008. His master thesis includes a novel hybrid electric vehicle control strategy and wheel-hub motor design with the support of Energy Systems Research Center (CESR) in TTU. His research is on the design and control of the electric motors, hybrid electric vehicle applications, power system control and smart grid with distributed energy resources. He graduated from Istanbul Technical University with a BSEE in 2008 and recently participated in Higher Education Research Experience program in Oak Ridge National Laboratory (ORNL), Oak Ridge, TN. Burak Ozpineci (S'92 M'02 SM'05) received the B.S. degree in electrical engineering from the Middle East Technical University, Ankara, Turkey, in 1994, and the M.S. and Ph.D. degrees in electrical engineering from the University of Tennessee, Knoxville, in 1998 and 2002, respectively. He joined the Post-Masters Program with the Power Electronics and Electric Machinery Research Center, Oak Ridge National Laboratory (ORNL), Knoxville, TN, in 2001 and became a Full-Time Research and Development Staff Member in 2002 and the Group Leader of the Power and Energy Systems Group in 2008. Presently, he also has an Adjunct Faculty appointment with the University of Arkansas, Fayetteville and The University of Tennessee, Knoxville. He is doing research on the system-level impact of SiC power devices, multilevel inverters, power converters for fuel cells, PV, and hybrid electric vehicles, and intelligent control applications for power converters. Dr. Ozpineci is the Chair of the IEEE PELS Rectifiers and Inverters Technical Committee and was the Transactions Review Chairman of the IEEE Industry Applications Society Industrial Power Converter Committee. He was the recipient of the 2006 IEEE Industry Applications Society Outstanding Young Member Award, 2001 IEEE International Conference on Systems, Man, and Cybernetics Best Student Paper Award, and 2005 UT-Battelle (ORNL) Early Career Award for Engineering Accomplishment.