Plug-in electric vehicles (PEVs) are here, and more are coming. In mid-2013, a buyer took delivery on the 100,000th PEV sold in the United States, and the number is growing. One study forecasts that more than a million plug-in hybrid electric vehicles (PHEVs) will be sold in California, New York, Washington, and Florida alone between 2013 and 2022. Electric vehicles (EVs) are also growing in range and sales. As the cost of battery packs comes down, the number of car shoppers willing to consider buying EVs will go up.
The growth of the PEV fleet means that an unplanned but potentially valuable energy storage resource is also growing—the battery packs of these vehicles. When PEVs are plugged in, they represent an opportunity to better manage the electricity grid. For instance, PEVs can be used to avoid potential shortages of electricity during peak times, provide extra storage capacity when the grid is generating more than it needs to satisfy demand, and encourage the growth of renewable energy by providing a buffer to balance out the intermittency of wind and solar generation.
Providing these vehicle-to-grid (V2G) services can also help the automotive market. The revenue from using a PEV to provide energy services to the grid could help offset some of the increased cost of purchasing the PEV. The increased incentive to buy PEVs replaces high emissions vehicles with no or low emissions vehicles. This shifts air pollution away from population centers, helps meet increasingly stringent emissions regulations, and assists car manufacturers with meeting future CAFE (corporate average fuel economy) standards.
There is a need for proven technologies that can predict the grid availability of a collection of independently operated vehicles. Yet the electricity grid needs to precisely match the demand for power from second to second with supply, drawing on a variety of sources ranging from base and peaking power plants to intermittent sources such as wind and solar power.
How can electricity grid managers, government authorities, power markets, entrepreneurs, and other stakeholders harness the resource offered by the growing fleet of PEVs? A project at the Environmental Energy Technologies Division of Lawrence Berkeley National Laboratory (Berkeley Lab) may help to answer that question.
“Many are saying that energy storage is important to the electricity grid, because it can be a buffer for the grid by providing power when it needs to smooth out sudden increases in demand or shortfalls in supply, or by storing power when an excess is available on the grid. Energy storage is high-value if it is able to respond quickly. The battery packs in electric vehicles are potentially very quick, compared to conventional sources today,” says Samveg Saxena, a researcher in EETD’s Grid Integration Group. When plugged into the grid, these vehicles could be a significant resource.
But there are many uncertainties that make using PEVs difficult: at any time, some PEVs are parked and charging up from different states of power depletion, and others are in use on the roads, so the capacity of the PEV fleet is always changing. Despite the opportunities, there are still many uncertainties, such as the effects upon battery degradation and battery lifetime from storing and sending power to and from the electricity grid, or whether enough PEVs can be tapped exactly when they are needed to meet grid demand.
Beyond this, says Saxena, “automotive and electricity utility stakeholders have not historically had to deal with these challenges together. Automotive battery manufacturers have to make sure that these grid services won’t degrade the batteries. Electric grid operators have to make sure that PEVs can function effectively as a grid resource.”
To study these issues, Saxena, EETD researcher Jason MacDonald and UC Berkeley/EETD Professor Scott Moura have been developing a simulation platform called the Vehicle-to-Grid Simulator, or V2G-Sim.
“V2G-Sim’s purpose is to be a simulation platform that couples sub-modules that address these concerns in a systematic way. It will help us understand the challenges of vehicle-to-grid services as well provide a platform for thinking through solutions, and simulating the effect of those solutions on the grid quantitatively,” he explains.
The team’s goal for V2G-Sim is to provide a platform for electric grid system operators, utilities, policy-makers, battery and PEV manufacturers, researchers, and the business community. Each stakeholder may study and evaluate their perspectives on utilizing PEVs for energy services to the electric grid (one example is what the utility community calls ancillary services).
V2G-Sim models the usage of individual vehicles, including second-by-second energy use while driving or charging, and aggregates large numbers of simulated vehicles to produce grid-scale predictions of impacts and opportunities from vehicle-grid integration. The results are time-based models of vehicle behaviors as well as a spatial simulation of their location. Using the National Household Travel Survey, the development team has created profiles of vehicles approximating real-life situations. For example, it could emulate a car that charges overnight, leaves for work at 7:30AM, parks, runs an errand at lunchtime, and then drives home at 5:30pm and plugs in to recharge. This is one of many scenarios modeled with statistical variations derived from real-world commuting data.
The preliminary version of V2G-Sim that the EETD team has created incorporates modules that address different aspects of the problem. Powertrain modules calculate the vehicles’ states of charge and energy use second-by-second. Battery electrochemistry modules calculate the electricity inputs and outputs and changes to their internal chemistry. Battery degradation models integrated into V2G-Sim estimate the impact of battery use on its life—how many years it can last when being used for driving only versus driving plus grid services.
Figure 1 shows the results of a test case with V2G-Sim—how demand from the grid is using electricity from 1,000 PEVs second-by-second over 24 hours starting at midnight. Figure 2, from the same test, shows the activity profiles of 12 individual vehicles—their states of charge as a function of time. Sometimes they are plugged in and charging, other times, they are unavailable. Figure 3 shows an example of the spatial resolution from V2G-Sim. In this example spatial charging is resolved for 659 PEVs at home and work locations in the San Francisco Bay Area, but V2G-Sim enables much finer spatial resolution of vehicle charging, for example by neighborhood.
Figure 1. V2G-Sim test case results—grid demand for electricity from 1,000 PEVs over 24 hours.
Figure 2. V2G-Sim activity profile for 12 vehicles.
Figure 3. Spatial charging for 659 PEVs at home and work locations in the San Francisco Bay Area.
“In the near term,” says Saxena, “our vision is to release ‘V2G-Sim Analysis’ as a research tool to improve the cross-disciplinary understanding of how V2G services could perform, the impact that vehicle-grid integration will have on individual vehicles and on the grid, and how grid infrastructure can be planned for more PEVs.” V2G-Sim Analysis will provide a valuable research tool for many parties to quantitatively understand the challenges from vehicle-grid integration, including grid operators, utilities, policy makers, battery manufacturers and the business community. But they plan to follow up with another version of the platform, ‘V2G-Sim Operations’, to enable real-time operations of a grid, which uses many PEVs as a resource. “The approach we’re taking,” he says, “is to develop a tool that will have a broad impact, and to ramp up the real-time use of PEVs to provide rapid energy response services to the grid.”
In this vision, battery manufacturers might use V2G-Sim Analysis to link to their electrochemical models of battery technologies to quantify battery degradation and devise ways of making long life-cycle batteries that effectively provide both vehicle propulsion and electric grid services. Advanced battery technology researchers at Berkeley Lab and elsewhere are already interested in using V2G-Sim in their studies of the electrochemistry of battery degradation. PEV manufacturers could link the platform to their own vehicle design platforms to adapt powertrain design for electric grid integration.
Grid managers could eventually use the Operations version of the platform to coordinate PEV resources in real-time for grid services such as smoothing the electricity supply curve from the intermittence of renewables such as wind and solar power. This, in turn, might help encourage greater use and integration of renewable power sources on the grid, because its managers have a greater ability to compensate for power fluctuations from these sources.
Utilities and business entrepreneurs could use V2G-Sim Analysis to develop and understand the impact of managed charging control algorithms for PEVs so that they can provide optimal service both as a vehicle and as an energy service to the grid.
Entrepreneurs interested in creating a business by harnessing the fleet of PEVs of a region with a large PEV stock could use V2G-Sim to make informed decisions about how much regulation capacity to bid into the grid’s markets. Information such as the composition of the PEV fleet, the state of charge and availability of energy at any time of the day or night, the locational availability of power throughout the region—where PEVs are connected to the grid—all influence these decisions.
The electricity regulatory community can use the same type of information to integrate the regulation of V2G services into the current regulatory framework of the grid.
“Our short-term goal,” says Saxena, “is to release V2G-Sim Analysis to the research community, automotive and battery manufacturers and grid stakeholders. In the long term, we want to further validate the model so that system operators and the regulatory community can begin to use PEVs in realtime as part of the dynamic electricity system.”
For more information, contact: Samveg Saxena, SSaxena@lbl.gov
This research was funded by Berkeley Lab’s Lab-Directed R&D program.