Technology 3 min read

Using AI to Help Drive Engine Efficiencies On The Road

Image courtesy of Shutterstock

Image courtesy of Shutterstock

The researchers at Argonne are developing a deep learning framework that could increase engine efficiencies.

Along with reducing emissions, the new model will provide improved engine performance and fuel economy. The researchers dubbed the new project as the Machine Learning Tool for Engine Simulations and Experiments or MaLTESE for short.

Car buyers now expect more from their vehicles. Aside from the usual demand for better engine performance and fuel economy, automotive manufacturers now face the pressure of building cars with low emission. Meeting these ever-increasing demand is no easy task.

However, the researchers at the U.S. Department of Energy’s (DOE) Argonne National Laboratory‘s deep learning framework will enable manufacturers to do just that – meet the demands.

With access to the supercomputing resources at Argonne, researchers Shashi Aithal and Prasanna Balaprakash started developing MaLTESE.

The framework is supposed to provide a manufacture-like onboard system that combines the power of high-performance computing (HPC) with machine learning. This will enable a new generation of autonomous cars with real-time adaptive learning and controls.

But to get to that stage, the team had first to understand how thee diverse driving and engine operating conditions affect engine performance and emissions.

In a statement, Balaprakash said:

“For the given driving conditions and driving behavior, we want to know a multitude of things, like nitrogen oxide and carbon emissions, and efficiency.”

Simulating Driving Conditions to Improve Engine Efficiencies

Balaprakash and Aithal used MaLTESE to simulate a typical 25-minute drive cycle of about 250,000 vehicles. That’s nearly as much traffic flow as you’ll find on four major highways in Chicago during rush hour.

By using Theta system—one of the world’s most powerful supercomputers— to nearly a full capacity, the team was able to complete the simulation in less than 15 minutes.

Completing such a high-fidelity simulation of just one engine cycle usually take several days, even on a supercomputer. That’s because a typical commute has thousands of different engine cycle.

However, Aithal had previously developed a physics-based real-time engine simulator called pMODES (parallel Multi-fuel Otto Diesel Engine Simulator).

The simulator runs faster than traditional engine modeling tools. Also, it can concurrently simulate the performance and emissions of thousands of drive cycles.

MaLTESE is a merger of the simulation-driven deep-learning tools being researched by Balaprakash and Aithal’s pMODES.

The researchers used the engine simulation outputs from pMODES to train a deep neural network to learn driving conditions and engine designs affect the vehicle’s performance and emissions. Using this information, the neural network can then predict the engine performance and emissions for a set of inputs in microseconds.

This brings us closer to within the realm of onboard real-time adaptive control.

Aithal noted:

MaLTESE could lead to a rapid paradigm shift in the use of HPC in the design and optimization and real-time control of automotive-features with far-reaching implications for autonomous and connected vehicles.”

Read More: New Artificial Intelligence Technique to Forecast Volcanic Eruptions

First AI Web Content Optimization Platform Just for Writers

Found this article interesting?

Let Sumbo Bello know how much you appreciate this article by clicking the heart icon and by sharing this article on social media.

Profile Image

Sumbo Bello

Sumbo Bello is a creative writer who enjoys creating data-driven content for news sites. In his spare time, he plays basketball and listens to Coldplay.

Comments (0)
Most Recent most recent
share Scroll to top

Link Copied Successfully

Sign in

Sign in to access your personalized homepage, follow authors and topics you love, and clap for stories that matter to you.

Sign in with Google Sign in with Facebook

By using our site you agree to our privacy policy.