The projects mark the fourth round of awardees under the HPC4Manufacturing (HPC4Mfg) programme since it was initiated in 2015. The programme, led by Lawrence Livermore National Laboratory (LLNL), Oak Ridge National Laboratory (ORNL) and Lawrence Berkeley National Laboratory (LBNL), teams DOE's national laboratories with U.S. manufacturers to address complex challenges with high-performance computing, including improving energy efficiency, reducing emissions, and maintaining American competitiveness. In addition to projects at the three leading laboratories, HPC4Mfg includes projects with Argonne National Laboratory (ANL), the National Renewable Energy Lab (NREL), the National Energy Technology Lab (NETL) and Sandia National Laboratories (SNL). Sandia is a new participant in the programme.
"We are excited that HPC is gaining traction in the manufacturing section, as evidenced by the large number of proposals received in this solicitation - a 50 percent increase compared to previous solicitations", stated Lori Diachin, director of the HPC4Mfg programme. "This led to an outstanding set of funded projects."
The seven projects awarded funding under the latest HPC4Mfg round include: LLNL partnering with Vitro Flat Glass to develop a neural network model for glass furnace operations; LBNL collaborating with PPG Industries to optimize rotary bell atomization; ORNL working with Eaton Corporation to enable next-generation low-temperature waste heat recovery; and SNL partnering with Vader Systems on computational modelling of liquid-metal 3D printing.
For the fifth project, ANL will partner with Caterpillar to optimize heavy-duty diesel engines for reduced emissions and improved fuel economy. The sixth project is led by General Motors, who will partner with LLNL to develop new models of high-pressure resin transfer molding to reduce the cost of manufacturing automotive structural carbon composites. Both projects are co-funded by the AMO and the Vehicle Technology Office (VTO) in DOE's Energy Efficiency and Renewable Energy Office, which is putting $350,000 toward the efforts.
The seventh project is led by Arconic, who will work with LLNL and ORNL to improve microstructure modelling of additively manufactured metal parts. The project is funded jointly by DOE's Office of Fossil Energy and the AMO, and is one of the first projects under the new HPC4Materials programme, aimed at accelerating industry discovery, design and development of new or modified materials for severe environments.
The Advanced Manufacturing Office within DOE's Office of Energy Efficiency and Renewable Energy funded LLNL to establish the HPC4Mfg Programme in 2015. To date, the programme has provided approximately $15 million in funding for a total of 47 private-public projects, ranging from improving energy efficiency in paper making and automobile engines to improving the reliability of wind turbines and reducing defects in 3D-printed parts. Under the program, industry partners provide in-kind contributions of at least 20 percent of the total project budget. The next round of solicitations will take place in early 2018.
The compute resources for this programme are provided through a variety of mechanisms, including the Advanced Scientific Computing Research Program within DOE's Office of Science Leadership Computing Challenge allocation program, allocations at NREL, NETL and other NNSA laboratories.
The full list of projects is available at the HPC4Mfg website.