Newswise – A new way of looking at artificial intelligence can open many doors for you 3D. Printing and construction of modern nuclear reactors.
The future of clean energy is hot. Beat temperatures 800 Celsius in parts of solar power plants and modern nuclear reactors. It’s difficult to find materials that can withstand this type of heat. Experts are therefore looking for answers to Mark Messner.
A senior mechanical engineer with the U.S. Department of Energy (DOE) Argonne National Laboratory, Messner is part of a group of engineers looking for better ways to predict the behavior of materials under high temperatures and pressures. Current forecasting methods work well, but they take time and often require supercomputers, especially if you already have a number of specific material properties – e.g. B. Rigidity, density or strength – and want to find out what kind of structure a material needs to match these properties.
“Normally you would have to do tons of physics-based simulations to solve this problem, ”said Messner.
While looking for an abbreviation, he found that neural networks, a type of artificial intelligence (AI), which reveals patterns in huge data sets, can accurately predict what will happen to a material under extreme conditions. And that much faster and easier than standard simulations.
Messner’s new method found the properties of a material more than 2,000 Times faster than the standard approach, as reported in an October report 2019 Article in the Journal of Mechanical Design. Many of the calculations, Messner noted, could be performed on a normal laptop with a graphics processor (GPU) – instead of a supercomputer, which is often inaccessible to most businesses.
This was the first time anyone had used what is known as a convolutional neural network – a type of neural network with a different, simpler structure that is ideal for recognizing patterns in photos – to accurately identify the structural properties of a material. It is also one of the first steps in accelerating the development and characterization of materials by researchers, which could help us achieve a completely clean energy economy.
Cats on the internet play a role
Messner started as a postdoctoral fellow at DOELawrence Livermore National Laboratory, where a team tried to build structures on one 3D. Printers in the micrometer range or millionths of a meter. Although cutting edge, research has been slow. Could AI Speed up results?
Back then, technology giants in Silicon Valley had started using convolutional neural networks to recognize faces and animals in images. That was what inspired Messner.
“My idea was that the structure of a material should not differ from that of an a. differs 3D. Picture, ”he said.“It makes sense that the 3D. Version of this neural network will recognize the properties of the structure well – just as a neural network learns that a picture is a cat or something else. “
To test his theory, Messner took four steps. He:
- designed a defined square with bricks – like pixels;
- took random samples of this design and used a physics-based simulation to create 2 Millions of data points. These points connected his design to the desired properties of density and rigidity;
- fed 2 Millions of data points into the convolutional neural network. This trained the network to look for the right results;
- uses a genetic algorithm, a different type of AI designed to optimize the results together with the trained convolutional neural network to find an overall structure that corresponds to the desired properties.
The result? The new AI Method found the right structure 2,760 times faster than the physics-based standard model (0.00075 Seconds vs. 0.207 Seconds).
New tools promote nuclear innovation
This abstract idea could transform the way engineers design materials – especially those that are designed to withstand conditions of high temperatures, pressures, and corrosion.
Messner recently joined a team of engineers from Argonne and DOE‘s Idaho and Los Alamos National Laboratories to partner with Kairos Power, a nuclear startup. The team is developing AI-based simulation tools that will help Kairos design a molten salt nuclear reactor that uses molten salt as a coolant, unlike current reactors. With these tools, the team is called projecting, like a certain type of stainless steel 316H, will behave in extreme conditions for decades.
“This is a small but important part of the work we do for Kairos Power, ”said Rui Hu, a nuclear engineer who leads Argonne’s role on the project.“Kairos Power wants very accurate models of the behavior of reactor components within its reactor to support its license application to the Nuclear Regulatory Commission. We look forward to making these models available. “
Another promising avenue for this type of job is this 3D. To press. In front 3D. The pressure prevailed, engineers struggled to build structures like Messner’s AI in his 2019 Paper. Another structure layer by layer with a 3D. Printer allows more flexibility than traditional manufacturing methods.
The future of mechanical engineering can lie in the combination 3D. Printing with new AI-based techniques, Messner said.“You would give the structure – determined by a neural network – to someone with a. give 3D. Drucker and they printed it out with the properties we wanted, ”he said.“We are not quite there yet, but that is the hope. “
This research used the Argonne bebop cluster in its Laboratory Computing Resource Center.
Argonne National Laboratory seeks solutions to pressing national problems in science and technology. As the country’s first national laboratory, Argonne carries out leading-edge basic and applied scientific research in virtually all scientific disciplines. Argonne researchers work closely with researchers from hundreds of corporations, universities, and federal, state and local agencies to help them solve their specific problems, advance America’s scientific leadership, and prepare the nation for a brighter future. With employees from more than 60 Nations, Argonne is administered by UChicago Argonne, GMBH for the US Department of Energy’s Office of Science.
The Department of Energy’s Office of Science is the largest single funder of basic science in the United States, working to address some of the most pressing challenges of our time. Further information can be found at https: // ener gy .gov / s c ience.