An influx of electric trucks is expected on our highways, but where will they charge? Few public stations today can serve large freight vehicles. The challenge? Fast chargers are expensive, and new ones need to be placed strategically to keep pace with growing demand.
To address this challenge, a research team that included Will Northrop created a model-based, flexible framework that will help identify locations and optimize the charging infrastructure.
“Placing charging stations efficiently is critical to save both time and money while supporting expected demand from widespread vehicle electrification,” says Northrop, a professor in the U of M Department of Mechanical Engineering (ME). “Our innovative framework can optimize multiple objectives such as financial costs, wait times, and route viability.”
Within their new framework, the researchers conducted several rounds of driving simulations, modeling electric vehicles driving along a route. Vehicle battery levels were tracked using a simplified vehicle model. During the first round of simulations, the team allowed vehicles to charge at any point along the road network to assess areas of high charging demand.
Next, charging stations were placed in the model based on the results of the first simulation, and a second simulation was run with fixed charger locations to determine how many chargers are needed at each station. The framework assesses the effects charger-by-charger: It places one charger at a time and repeats the process of estimating demand, placing another charger, and assessing the effect of that placement. “In this way, the framework iteratively assesses the effects of each individual charger placement on financial costs and fleet electrification potential,” says Matthew Eagon, an ME graduate student on the research team. This process was applied to the real world with a set of simulations that placed electric truck charging stations along five major highway corridors used frequently by long-haul trucks.
Using the real-world scenarios, the team developed a model to optimize charger placements based on the expected charging demand. The model sets limits on expected wait times at each station to ensure that an appropriate number of chargers are placed to accommodate times of peak demand. The framework is set up to allow vehicle, charger, and optimization models to be swapped out and compared. “The results demonstrate the flexibility and potential effectiveness of the new framework for scalable charger-station deployment,” Northrop says.
In the future, researchers envision using driving data collected from fleets of cloud-connected vehicles to generate results. “Eventually, this could lead to an open-source tool where other researchers could analyze the effects of changing the optimization objectives and adding more information to increase the nuance of the optimization,” Northrop says.
The analysis is based on work supported by the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy under the Vehicle Technologies program, Award Number DE-EE0008805.
It was published in a paper titled Model-Based Framework to Optimize Charger Station Deployment for Battery Electric Vehicles in the 2022 IEEE Intelligent Vehicles Symposium (IV) proceedings.
Writers: Megan Tsai and Pam Snopl
