As the number of plug-in electric vehicles (PEVs) in the United States quickly grows, new challenges and opportunities are emerging as electricity demand increases. These challenges compound the existing complexity of factors affecting metropolitan transportation networks, from traffic flow to emissions.
To accelerate and support the increased adoption of PEVs, cities need to provide new public charging station options in addition to at-home and at-work infrastructure. Multiple charging levels, including fast charging, will also be required, to enable the use of those vehicles for a wide range of usage (e.g., taxi, ride-hailing, car-sharing and long trips).
When drivers decide to charge is also very important for the electrical grid. Drivers will want to charge their vehicles not only during the day, but also at home. Without proper management, demands on the electrical grid may spike, straining the system. Understanding the demands at the individual vehicle level will help utility companies and city planners decide when and how to best meet and/or modify those demands. It will also guide them as they adapt and improve their services and electrical infrastructure with technology and policy options.
These are just a few of the challenges to consider when analyzing the impacts of these new vehicle technologies on large-scale transportation networks.
At Argonne, scientists in the Energy Systems Division have combined several of Argonne’s modeling and simulations capabilities to meet these challenges. The result is a new tool coupling Polaris, an open-source transportation systems model, with the Argonne Least-Cost Energy Framework (ALEAF), an electrical grid model to simultaneously model the electricity demand and supply.
This large-scale modeling of a transportation network’s energy usage is made possible by the sheer scale of POLARIS’s simulation capabilities. POLARIS uses a discrete-event, agent-based modeling framework to simulate the behavior of individual vehicles within any given transportation network.
“Each agent makes its own travel decisions, from departure times and route choices to destinations and on-the-fly rescheduling,” said Systems Modeling and Control Section Manager Aymeric Rousseau. “This kind of integration and simulation of traveler decision-making behavior establishes Polaris as truly one of a kind.”
The ability to model energy usage at the level of millions of different agents representing specific vehicle configurations allows for a much higher fidelity of modeling. “While most other tools use constant values for energy consumption per hour, Argonne uses a machine learning model to take into account the variance in electrical consumption rates across different vehicles at different times of the day,” explained Principal Computational Transportation Engineer Omer Verbas. “This gives us better estimations of energy consumption, and thus better estimations for charging station and grid requirements.”
The electricity demand from individual vehicles is then provided to ALEAF. “With ALEAF, we can manage how the electricity is provided to the vehicles but also how the vehicles can help make the grid more resilient, for a two way approach,” explained Principal Computational Scientist Zhi Zhou.
The new capability allows Zhi and Verbas to examine various elements of this energy consumption. “We can look at how the grid dictates charging needs and options and vice versa,” said Verbas. “For example, we can analyze when optimal charging times would occur based on the needs of the grid or the needs of the vehicles.”
Rousseau further explained that they can use these simulation iterations to determine the best charging times or availability of renewable energy on any given day.
“You do not want to cripple the grid if everyone plugs their car in directly after they get home from work,” Zhi explained. “You would need to stagger charging schedules in order to better support the electrical infrastructure, and to optimize the system.” Having a robust research platform to give these detailed insights leading to decisions, is far superior to gut-based estimation methods.
Zhi and Verbas are also working with Argonne’s Smart Energy Plaza, a remodeled gas-station-turned-advanced charging-station research platform and microgrid, designed to study electric energy usage and battery life of electric vehicles. The plaza helps quantify when drivers are charging their electric vehicles and how often they are using fast chargers. With this data, scientists can examine the impacts that a transportation network comprised of more electric vehicles would have on the battery life of these vehicles under various usage cases.
“By combining these capabilities here at Argonne,” described Rousseau, “we are able to examine everything from batteries, to vehicles, to city infrastructure, to entire electrical grids. They all come together in a system of systems that characterize the energy consumption in these large-scale transportation networks.”
The work of Polaris with ALEAF and Argonne’s Smart Energy Plaza is funded through the U.S. Department of Energy’s Laboratory Directed Research and Development.