GM, National Labs Paving the Way for Next Generation Vehicles

For the better part of a century, General Motors (GM) was the world’s largest automaker. Now, amid a paradigm shift toward smarter electric cars, the leading US automaker is seizing the moment — and to do so, it is tapping into deep partnerships with US National Laboratories, from particle accelerators to their supercomputers. At a meeting of the Department of Energy’s Advanced Scientific Computing Advisory Committee (ASCAC) last week, Paul Kragowski — director of vehicle systems research at GM’s Research and Development Center — highlighted the depth and value of GM’s work with national laboratories to leverage HPC and advanced vehicle technology. .

Covers many national laboratory partners HPCwire Finding the need for national laboratories only when they reach the limits of their computational resources. But GM, Krajewski explains, fosters deep relationships with national laboratories that permeate many aspects of its research and development — not only through its high-performance computing equipment, but also through experimental equipment, research expertise and software.

Krajewski noted that National Laboratories’ collaboration with GM spanned many forms: independent projects, contracts, partnerships, mutual participation in broader DOE contracts, facility use, discussions, and paper publications. Through these collaborations, he said, GM has worked to capitalize on “excellent researchers” and “unique capabilities” in national laboratories — with an emphasis on data collection.

“One of the opportunities that I think is really important for this collaboration is shared access to data,” he said. “Computing doesn’t give us the right answers if we don’t have the data – if we don’t have the ground truth and we don’t have the empirical results.”

Image courtesy of General Motors.

Krajewski provided a high-level overview of where GM is investing its research money: “Battery Technology, Autonomy, Fuel Cell and Advanced Manufacturing,” with “nearly half” of the funding allocated to battery and electrical technology.

“We are committed to an electric future,” Krajewski said. He explained that GM has partnered with National Laboratories through a consortium called Battery500. The consortium — which operates in four national laboratories (led by Pacific Northwest National Laboratory) and five universities — aims to develop high-energy, rechargeable lithium-metal batteries for electric vehicles. (Lithium-metal batteries differ from lithium-ion batteries in that they offer significantly higher energy density but suffer from stability challenges.)

Through the Battery500, GM has worked with national laboratories to develop new processes and materials for lithium-metal batteries, informing subsequent simulations and significantly developing models. “This is really important in generating the experimental data we need to build the models,” Krajewski said. “There is also work now on multi-scale modeling of lithium-ion batteries, and then, eventually, you’ll go beyond that to lithium-metal batteries, and that’s where the computational ability is really important. If you want to be able to do these multi-physics models, Multiscale that is so computationally intensive, you need to have that power that you have in national laboratories.”

But GM doesn’t put all its eggs in the battery’s electric basket — much of GM’s other research focuses on fuel cell technology, which might use hydrogen fuel cells to power electric motors, with only water and heat as by-products. Krajewski said there is a “tremendous amount of work going on” with the National Renewable Energy Laboratory (NREL) on fuel cells, given both experimental capacity and computer modeling. For example, General Motors helped fund the H2FiLLS hydrogen fuel cell filling simulation project, which was making good use of the Intel-powered Eagle supercomputer (8 petaflops peak) to run computational fluid dynamics models of hydrogen storage tanks, mixing simulation results with Beta tests to further improve the simulation.

Similarly, labs and General Motors are working together on the thermal management of electric vehicles. “Whether it’s thermal management of the batteries themselves or the electric motors, it becomes a very important challenge, and so there is some work now going on analyzing thermal management with Argonne,” Krajewski said. [National Laboratory] On the engines and how [we can] Optimizing pushing these motors to their peak performance…without causing issues with overheating and how to manage the thermal stresses we end up building in these motors while we push them to the limit. This work is supported by the allocation of 5 million base hours on the Intel-powered Argonne Theta platform (6.9 Linpack petaflops).

Manufacturing and lightweight are also major areas of GM’s focus. “When you look at the data that we have on manufacturing capacity in the future, when you think about moving into chip production, mining, critical materials — all of these things are going to require manufacturing models so that we can be very efficient at making them very efficient,” Krajewski said.

On the lightweight front, GM is working with the Advanced Photon Source (APS) Accelerator at Argonne National Laboratory to work on a characterization of so-called “Generation 3” steels, which are intended to be stronger (and thus require fewer materials for adequate performance and safety in the vehicles). These experimental results from APS were then combined with finite element optimization on NREL’s (now decommissioned) Peregrine array. GM similarly partnered with Oak Ridge National Laboratory (ORNL) to develop a new aluminum alloy (DuAlumin 3D) through a combination of computational and experimental analysis.

Advanced Photon Source. Image courtesy of Argonne.

“How can I use this material once I have developed it?” Kragowski continued. “Now I have a set of materials that I can apply; I have performance requirements – in this case, … a body profile with different parts; I can run finite element optimization using block in NREL to improve on materials, metrics, where I want to use it, and really come up with To the best solution for applying these materials I also designed to take advantage of these technologies.”

Then, even later in the design process, GM “targets 100% virtual verification by 2025” to eliminate unnecessary molds, tools and prototyping parts and speed up vehicle development. “When we have those physical models, now [you] I can do things like predict performance and predict shaping potential – can I make the shapes I want to make? – Principles of static and dynamic, collision performance.

“This cooperation is necessary to advance technological development,” Krajewski concluded. “There are a lot of models for this collaboration, and combining the ability of experiments with this computational ability is key to this collaboration and really moving the technology forward.”

Leave a Comment