The Lawrence Livermore National Laboratory (LLNL) is a hub for scientific and technological discovery. It is also home to 14 supercomputers, more than any other known computing institution in the world. LLNL was founded in 1952 to advance nuclear weapons science and technology. At the time, its foremost goal was to develop ways to protect American national security. While that is still a critical part of LLNL’s mission, the lab has also developed different ways to collaborate with industry and private companies to advance broader scientific and technological discoveries.

Many Business Opportunities

Livermore Valley Open Campus, for example, is a unique partnership between Lawrence Livermore National Laboratory and Sandia National Laboratories that allows industry and academic collaborators to tap into the resources of both facilities, which use some of the world’s most advanced science and technology. This program lets industry collaborate with lab personnel on research that often leads to new patents, products, and technology.

There are several other LLNL programs that offer commercial enterprises access to the deep expertise and cutting-edge equipment available through the lab. LLNL’s Innovation and Partnerships Office (IPO) “serves as a focal point for LLNL engagement with industry,” lab officials say. “Whether by technology commercialization, encouraging entrepreneurship, or via laboratory business development activities, our mission is to grow the economy by advancing the development and commercialization of scientific discoveries. Our goal is to identify new economic opportunities and solutions and transfer those to the private sector through licensing or partnerships for the benefit of the U.S. economy.”

Another program that supports collaboration with private companies is High Performance Computing for Energy Innovation (HPC4EI). The program was created by upper management at LLNL who believed that industry could benefit from access to high performance computing resources. With funding from the U.S. Department of Energy (DOE), the pilot program launched in 2015. It matched private companies with LLNL scientists and lab resources to help companies improve manufacturing processes, energy efficiency, and decarbonization.

After the successful pilot, “the DOE, national lab, and industry partners were all very happy with the results of the project,” says HPC4EI Director Aaron Fisher, who notes that 11 national laboratories participate in HPC4EI, which is managed by LLNL. Twice a year, DOE offices give program officials a list of manufacturing industries and specific technologies of interest. Then HPC4EI publishes a solicitation document outlining what it seeks. Companies are invited to submit short concept papers explaining how they would use access to national lab resources for up to one year to meet DOE goals.

“We’re looking for projects that will help companies save money while also saving energy and reducing carbon output for the nation,” according to Fisher. “From the national lab perspective we are taking all of this technology that we have built up to solve problems for the DOE and applying it to problems in industry such as making steel more efficiently and without pumping carbon into the atmosphere. The folks in the businesses are looking for more efficient processes so they can make more money, while the folks in the DOE are looking for energy and carbon savings. And the folks in the national labs are looking to apply their technologies more broadly in the world.”

The Power of Computing

High performance computing uses sophisticated technology to process data at dazzling speeds far faster than typical computer systems. Computational fluid dynamics, materials science and engineering, computational chemistry, and seismic imaging and analysis are among the applications made easier, or sometimes simply possible, by high performance computing. It increasingly powers the process of discovering new drugs as well as artificial intelligence developments.

“We’re experts at using these computers to solve large-scale problems,” Fisher says. “On the business side, companies have decades of expertise in their respective industries. We don’t have people with that deep expertise in making steel, aluminum, glass, cars, etc. These two communities learn to speak each other’s languages and work together over the course of the project. We hope that in addition to the results companies get out of this project, they really see the power of computing. Our overarching goal is to push this computing technology out into the manufacturing world.”

The HPC4EI program is an elegant approach to meeting the needs of all three stakeholders. Companies join the HPC4EI program to learn how to make their products more efficiently, less expensively, or higher quality. The DOE focuses on projects with significant energy and CO2 impacts. The national labs benefit because the program gives early career scientists opportunities to lead projects and solve problems on a broader scale.

As DOE officials explain HPC4EI, “selected industry partners are granted access to high performance computing facilities and world-class scientists at Department of Energy’s national laboratories. Through the initiative, our nation’s world-class supercomputers are helping manufacturers achieve significant energy and cost savings, improve product performance, expand their markets, and grow the economy.”

Essentially, HPC4EI acts as a kind of matchmaking service. If a concept paper seems promising but needs more work, Fisher’s team may recommend that a company work with a particular national lab scientist to help flesh it out. If a concept is accepted, a DOE national laboratory principal investigator is assigned to help the company develop a full proposal at no charge. The DOE typically funds eight to 10 proposals each period. It pays labs up to $400K for sharing access to lab resources and expertise with companies accepting into the program, while the companies pay a minimum of 20% of project costs either in cash or in-kind.

According to officials, LLNL worked with Seurat Technologies “to model a faster laser powder bed manufacturing process and bring it to market.” It also partnered with Materials Sciences “to deploy topology optimization, additive manufacturing, machine learning, and ultra-high-powered computers to optimize exchanger physics and slash compressor energy consumption by 30 percent.” Flash Steelworks “collaborated with Oak Ridge National Laboratory to utilize phase field simulations to gain fundamental understanding of local composition gradients during flash annealing” while Argonne National Laboratory and Raytheon Technologies Research Center developed “data-driven approaches for optimizing gas turbine film-cooling designs while reducing turbine cooling air, thereby improving engine efficiency without sacrificing component durability.” In short, the program has created a string of success stories for a variety of manufacturers and actively encourages small and medium-size companies to apply for the program.

“We are trying to get the U.S. industry to use this technology to further their goals,” notes Fisher. “High performance computing has been part of our game for decades. It's only just beginning to be part of the game in most industries. So we're trying to help them in their journey on this. We also host workshops and seminars. We are a pretty small program, and we can't move the entire country right away. But one company at a time, I feel like we are making a difference.”

For more information about Livermore Valley Open Campus, please visit lvoc.org.

For more information about the Lawrence Livermore National Laboratory Innovation and Partnerships Office, please visit ipo.llnl.gov.

For more information about the High Performance Computing for Energy Innovation program, please visit www.hpc4energyinnovation.llnl.gov, www.facebook.com/eeregov/videos/high-performance-computing-for-energy-innovation/316117573522026, or www.linkedin.com/company/lawrence-livermore-national-laboratory.

Photo by Garry McLeod/LLNL