首页|New Findings on Machine Learning from Virginia Polytechnic Instituteand State U niversity Summarized (Digital Twins for RapidIn-situ Qualification of Part Qual ity In Laser Powder Bed FusionAdditive Manufacturing)
New Findings on Machine Learning from Virginia Polytechnic Instituteand State U niversity Summarized (Digital Twins for RapidIn-situ Qualification of Part Qual ity In Laser Powder Bed FusionAdditive Manufacturing)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News – Investigators publish new report on Ma chine Learning. According to news reportingfrom Blacksburg, Virginia, by NewsRx journalists, research stated, “This work concerns the laser powderbed fusion ( LPBF) additive manufacturing process. Currently, LPBF parts are inspected post-p rocessusing such techniques as X-ray computed tomography, optical and scanning electron microscopy, amongothers.”Financial supporters for this research include National Science Foundation (NSF) , Office of NavalResearch, Naval Surface Warfare Center (NAVAIR), National Inst itute of Standards & Technology (NIST)- USA, National Institute o f Standards & Technology (NIST) - USA, Institute for Critical Tech nologyand Applied Science, Macromolecules Innovation Institute, Office of the V ice President for Research andInnovation.The news correspondents obtained a quote from the research from Virginia Polytec hnic Institute andState University, “This empirical build-and-test approach for qualification of part quality is prohibitivelyexpensive and cumbersome. To ena ble rapid and accurate in-situ qualification of LPBF part quality, in thiswork, we developed a physics and data-integrated digital twin approach. To demonstrat e the approach,Inconel 718 parts of various shapes were manufactured under diff ering LPBF processing conditions. Theprocess was continuously monitored using i n-situ thermal and optical tomography imaging cameras. Thepart-scale thermal hi story was predicted using an experimentally validated computational thermal simulation. The simulationderived thermal history and sensor signatures were used as inputs to a k-nearest neighbor machine learning model. The machine learning mod el was trained with ground truth porosityand microstructure data obtained from post-process characterization. The approach predicted the onsetof porosity, mel tpool depth, grain size, and microhardness with an accuracy exceeding 90 % (R-2).”
BlacksburgVirginiaUnited StatesNor th and Central AmericaCyborgsEmerging TechnologiesMachine LearningVirgin ia Polytechnic Institute and State University
NETL
NSTL
