Lightning Systems researching AI for vehicle-emissions predictions

LOVELAND — Lightning Systems, a leader in emissions efficiency solutions for commercial vehicles, is using a form of artificial intelligence to predict the fuel economy and tailpipe emissions on fleets.

Engineers from Loveland-based Lightning Systems are using artificial neural networks, which are computing systems made of highly interconnected processing elements. Those elements process information and can predict outcomes: In this case, they’re making highly accurate fleet management predictions.

The company has been able to accurately predict the fuel consumption and emissions of vehicles in commercial and government fleets using artificial neural networks.

SPONSORED CONTENT

How dispatchable resources enable the clean energy transition

Platte River must prepare for the retirement of 431 megawatts (MW) of dispatchable, coal-fired generation by the end of the decade and address more frequent extreme weather events that can bring dark calms (periods when there is no sun or wind).

“Our computer modeling demonstrates the accuracy of predictive analytics to help fleets manage fuel consumption, decrease their fuel usage and reduce emissions,” Tyler Yadon, director of analytics for Lightning Systems, said in a prepared statement. “The tools we are developing can take incredibly complex real-world problems and turn them into extremely accurate predictions about your fleet.”

Yadon oversees Lightning Systems’ predictive analytics program, LightningAnalytics, which helps fleets monitor vehicle maintenance and track routes.

Artificial neural networks can be beneficial because they’re less costly than existing simulation standards and are easier and faster to re-train and apply to new vehicles, said Brian Johnston, director of emissions regulation and strategy for Lightning Systems and the co-author of a paper on the subject to be published by Colorado State University.

“Customer drive cycles are not always reproducible for fuel consumption and emissions research,” Yadon said. “Our results indicate that artificial neural network models can be used for a variety of research applications due to their economic and computational benefits, such as improving vehicle-control strategies to reduce fuel consumption and emissions in modern vehicles.”

The research conducted by Colorado State University students and the findings on Lightning Systems’ artificial neural networks will be presented at the SAE World Congress in April.