A D V A N C E D M A T E R I A L S & P R O C E S S E S | A P R I L 2 0 2 1 6 MACHINE LEARNING | AI AI IMPROVES CHEMICAL SEPARATION Led by researchers at the University of Toronto (U of T) and Northwestern University, Evanston, Ill., a new study is using machine learning and artificial intelligence (AI) to create the best building blocks in the pool of materials used for a specific application, such as improving chemical separation in industrial processes. “We built an automated materialsdiscoveryplatformthatgenerates the design of various molecular frameworks, significantly reducing the time required to identify the optimal ma- terials for use in this particular process,” says U of T researcher Zhenpeng Yao. The team focused on development of metal-organic frameworks (MOFs) that are considered the ideal absorbing material for removing CO2 from flue gas and other combustion processes. “The new approach uses machine learning algorithms to learn from the data as it explores the space of materials and actually suggests new materials that were not originally imagined,” explains professor Randall Snurr of Northwestern. www.utoronto.ca. MACHINE LEARNING SUPPORTS AEROSPACE Using machine learning, researchers from the National Institute for Materials Science (NIMS), Japan, determined the optimum parameters for manufacturing high quality, Ni-Co-based superalloy powders at high yields—a promising development for aircraft engine materials. The team then demonstrated that these parameters led to low-cost manufacturing of powders suitable for high-pressure turbine disk production. Because metal 3D printing has been rapidly adopted in aerospace engine production, there is growing demand for the alloy powders these printing techniques require. When the materials are used to produce high-pressure turbine disks, they must be heat resistant, highly plastic, and homogeneous superalloy powders that can be processed into spheres. They also must be produced at high yields to reduce costs. Super- alloy powders are commonly produced for this purpose using large gas atomizers. The team used machine learning in an attempt to optimize gas atomization processes for the manufacturing of Ni-Co-based superalloy powders without relying on the knowledge of experts. As a result, the team succeeded in manufacturing fine-grained powders that can be processed into spheres. In addition, use of the parameters dramatically increased production yields from the conventional 10-30% to approximately 78% after performing experiments only six times without using previously collected data. The powder manufactured in this research was approximately 72% cheaper than commercially available powders when the prices of the rawmaterials were compared. The prediction accuracy of machine learning models increases as they receive more training data, and superalloy powder manufacturers possess largely unexploitedmanufacturing process data. Integrating this data may further improve the ability to predict optimum parameters, leading to even higher quality powders at lower cost. www.nims.go.jp/eng. Optimization of superalloy powder manufacturing processes using machine learning. Courtesy of NIMS. AI enables autonomous design of nanoporous materials. Courtesy of University of Toronto. Researchers at UCLA developed a new image autofocusing technique to digitally bring a given microscopy image into focus without the need for special equipment during the image acquisition phase. The approach is based on deep learning, where an artificial neural network is trained to take a single defocused image as its input to rapidly create an in-focus image of the same sample. ucla.edu. BRIEF
RkJQdWJsaXNoZXIy MTYyMzk3NQ==