Newswise – By integrating machine learning into the design process, ML-GA Dramatically speeds up computer-assisted virtual prototyping, reducing the product development phase from months to days while lowering computational costs.
The US Department of Energy (DOE) Argonne National Laboratory and Parallel Works, Inc., a Chicago-based company CHP software platform company, won the Federal Laboratory Consortium (FLC) Midwest Regional Award for Excellence in Technology Transfer for bringing Argonne’s machine learning and genetic algorithm (ML-GA) design of marketing optimization software.
This recognition marks the couple’s second great success. Argonne won a price of $750,000 three years ago from the DOEof the Office of Vehicle Technologies, within the Office of Energy Efficiency and Renewable Energy, through the Technology Commercialization Fund (TCF) program to integrate new features into ML-GA and make it more efficient and portable. It streamlines the process of integrating software with the Parallel Works business platform.
Current industry standards for design, which have long favored computer simulation over physical experimentation, remain remarkably slow. Advanced engines for automotive applications, for example, contain many design parameters that can become time consuming and expensive to optimize, even with computer modeling.
The ML-GA technology solves this critical problem.
By integrating machine learning into the design process, ML-GA Dramatically speeds up virtual prototyping based on computer-aided engineering simulation, reducing the product development phase from months to days – and, in doing so, also reduces computation costs.
Pinaki Pal, researcher at the Argonne Transport Research Center (CTR), led the effort for the lab.
“If you are able to design a product in a much shorter time frame, you also accelerate its development and its introduction to the market, âhe said.“It was a great success. Our ML-GA the technology can be incredibly useful for optimizing product design in a variety of markets.
Greg Halder, former scientist and current business development manager at Argonne, said he was delighted to see the lab’s efforts in AI honored by such a prestigious organization.
“It’s particularly amazing because we are recognized by our peers, âhe said.“This is an exciting new technology. We are just delighted. It all happened quite quickly, going from the discovery of Argonne software to commercial adoption in just a few years.
The technology, Pal said, marks a big step forward in product design.
The difference between ML-GA and the more traditional methods are that Argonne software learns simulation data adaptively. Scientists run simulations in small batches – called iterations – and train machine learning (ML) surrogate model on the simulation data. This surrogate model replaces the simulation itself within a genetic algorithm (Georgia) optimization pipeline: the overall execution time is drastically reduced by this much faster substitution model, which calculates objective functions very quickly.
Yet there are challenges to this approach. ML surrogate models may require a lot of simulation training data to achieve high accuracy. To get around this bottleneck, ML-GA uses a Super Learner framework that combines several ML algorithms to make predictions as good or better than those made by each individual ML algorithm. Additionally, an active learning technique is used to select the best possible design points to simulate during each successive iteration.
“This enables efficient exploration of large design spaces and significantly reduces the number of simulations required during the design process, âsaid Pal.
Argonne recently reported the benefits of these ML-GA automotive engine design optimization capabilities within a research article published in the renowned International Journal of Engine Research, in collaboration with Parallel Works, Inc. and Convergent Science, Inc.
ML-GA was recently licensed by Parallel Works, which was founded in 2015 by Michael Wilde, Matthew Shaxted and Michela Wilde. Parallel Works maintains close ties with the laboratory: it relies on open source CHP workflow automation technology developed by the lab with UChicago and the University of Illinois. And Michael Wilde was on entrepreneurial leave from the lab when he and his co-founders started their business.
“Our team is delighted to integrate the advanced technologies of Argonne ML-GA technology with high performance computing and make it accessible through our The learner works family of products for use in critical industrial applications and advanced industrial research, âsaid Wilde.
Shaxted can’t wait to see the results. Parallel Works hopes the new technology will improve the customer experience and add significant value to engineers in a wide variety of government and commercial manufacturing industries, including automotive design, heavy equipment design, consumer goods packaging. and hydrological engineering.
The company is also working with government and industry researchers to assess its use in ocean exploration, weather forecasting, and life science discovery.
“Our customers are increasingly focusing on realizing the economic benefits of machine learning technology to design improved products and provide advanced services, âsaid Shaxted, president of the company.“Parallel Works provides easy and scalable access to these benefits and is thrilled to be part of the team that receives this FLC honor.”
Suresh Sunderrajan, Associate Laboratory Director at Argonne’s Energy and Global Security Directorate, was delighted with the honor.
“Recognizing Argonne’s efforts to develop and commercialize this unique software technology, the FLC This award illustrates the success of our mission to have a positive impact on energy production, storage, conversion, distribution and efficiency, âhe said.
The laureates will be celebrated during the FLC Midwest & Southeast regional meeting held virtually from September 21–23.
The FLC was organized in 1974. More … than 300 federal laboratories, facilities and research centers and their parent agencies constitute the FLC community today.
The Office of Energy Efficiency and Renewable Energies supports early-stage research and development of energy efficiency and renewable energy technologies to enhance U.S. economic growth, energy security, and environmental quality.
Argonne National Laboratory seeks solutions to urgent national problems in science and technology. The country’s leading national laboratory, Argonne conducts cutting-edge fundamental and applied scientific research in virtually all scientific disciplines. Argonne researchers work closely with researchers from hundreds of businesses, universities, and federal, state, and municipal agencies to help them solve their specific problems, advance U.S. scientific leadership, and prepare the nation for a better future. With employees over 60 nations, Argonne is managed by UChicago Argonne, SARL for the United States Department of Energy Science Office.
The Office of Science of the United States Department of Energy is the largest proponent of basic physical science research in the United States and strives to address some of the most pressing challenges of our time. For more information visit https: // ener gy .gov / s c ience.