Energy labs build digital twin to model traffic congestion relief and energy savings
By digging into traffic congestion data for the area around Chattanooga, Tennessee, the Department of Energy aims to reduce vehicle energy use by up to 20% and recover up to $ 100 billion loss of productivity over the next 10 years.
Researchers from DOE’s Oak Ridge National Laboratory (ORNL) and the National Renewable Energy Laboratory (NREL) partnered with Chattanooga on simulations to determine which controls would ensure fuel efficiency while optimizing travel time, speed and driver safety.
To do this, the researchers harnessed 500 sensors – including automated cameras, traffic lights, vehicle GPS devices, radar detectors and weather stations – to build a digital twin of the city. The Tennessee Department of Transportation has sensors roughly every half a mile that collect volume and speed data for each lane of traffic, providing up-to-the-minute situational awareness, according to Jibo Sanyal, group leader. and senior scientist at ORNL’s Computational Urban Science Group.
The digital twin serves three purposes, he said: information about what’s going on right now, what could happen and what to do about it.
For example, if an intersection is saved in the middle of the day, researchers can use the digital twin to look for ways to alleviate those traffic jams. They can experiment with connecting to a small set of traffic lights to test algorithms that control signal scheduling, alternate routing, speed harmonization, ramp measurement, and dynamic speed limits.
“To be able to control something reasonably well, we have to know what’s going on and then be able to anticipate what may happen,” Sanyal said. “This is where this aspect of being able to connect to the detection infrastructure they have in the city is of utmost importance.”
By combining data with machine learning, researchers can estimate energy use and energy loss, identify which systems are losing energy and why, and model realistic alternatives to changing conditions and controls, NREL officials said.
“We’re basically trying to explore this unknown space, figure out how to ingest all these disparate sensors, build a situational awareness platform through the digital twin that could become the basis for effective control,” Rich said. Davies, Director of ORNL’s Sustainable Transportation Program. Many unknowns lie at the limits of all of these systems, he said. “Basic cellular communications that we can rely on, but how exactly you tell a car about the surrounding traffic situation, there isn’t really a standard for that.”
To handle the heavy analysis of real-time data calculations and how the changes affect the results, the team uses NREL’s Eagle supercomputer. It can perform 8 trillion calculations per second, allowing researchers to perform calculations in hours, minutes, or seconds.
“If they worked on a laptop, they would work for months,” Sanyal said.
Transportation operators, engineers and city managers can understand the results of the analysis using a visualization tool created by the team to present the data in the form of charts, graphs and animations. “We calculate various performance metrics, metrics, and differences as we make changes to the system, which gives us a good idea of how to measure improvements based on performance,” he said.
The project began in 2018 with a two-year award from the DOE, and the team recently renewed it for an additional two years to expand the field of study to include parts of Georgia, vehicles from freight and more features. Chattanooga Public Works and Georgia Department of Transportation joined Chattanooga and Tennessee DOTs and Chattanooga-Hamilton / North Georgia County Transportation Planning Organization in New Phase with Industry Partners Covenant Transport, FreightWaves and Siemens Mobility.
The DOE would ultimately like to see researchers achieve “a 20% energy savings across the region,” Davies said. With the growth of connected and automated vehicles and increasingly intelligent transportation infrastructure, the introduction of more active controls will allow managers to optimize traffic lights, routing and other infrastructure and benefit from improvements. to vehicle drive trains, he said.
However, such an optimized system is not just around the corner. He will need the support of local leaders.
“We were worried [about] how long it would take to get there because the level of connectivity is relatively low, the level of automation is very low, ”Davies said. “We believe the solution will likely require something that involves the people who operate the roads and who have the jurisdictional power to make changes to the system.”
The technology developed for Chattanooga is transferable to other places and other uses, such as the study of the interface between the electrical network and the electric vehicles to be recharged. “We believe that this type of situational awareness will help us plan for recharging in the future, knowing both the amount of traffic and the network situation,” Davies said.