首页|Findings from Chinese Academy of Sciences Provide New Insights into Nanocomposit es (From Processing To Properties: Enhancing Machine Learning Models With Micros tructural Information In Polymer Nanocomposites)
Findings from Chinese Academy of Sciences Provide New Insights into Nanocomposit es (From Processing To Properties: Enhancing Machine Learning Models With Micros tructural Information In Polymer Nanocomposites)
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-Investigators discuss new findings in Nanotechnol ogy - Nanocomposites. According to newsreporting originating from Changchun, Pe ople's Republic of China, by NewsRx correspondents, researchstated, "For polyme rs and their composites, processing conditions and the resultant microstructures arecrucial in determining their properties. Traditional machine learning (ML) methods typically focus onestablishing direct relationships between processing parameters and material properties, often overlookingthe critical intermediate step of how processing influences microstructure, limiting the predictive accuracy."Funders for this research include National Key R&D Program of China , National Natural ScienceFoundation of China (NSFC), Network and Computing Cen ter in Changchun Institute of Applied Chemistry.Our news editors obtained a quote from the research from the Chinese Academy of Sciences, "In thisstudy, we introduce an approach that first establishes a deta iled relationship between processing parametersand the resultant microstructure , and then uses transfer learning and feature fusion to integrate thisrelations hip into the prediction of material properties. Using carbon black-reinforced ru bber composites(CRC) as an example, we compared ML models in predicting mechani cal properties from processingdata. A multi-task deep neural network performed best achieving an R-2 of 0.763 with only processingdata as input. When incorpor ating transfer learning and feature fusion, the R-2 improved to 0.852 and0.878, respectively. Shapley explanation analysis validated our approach, highlighting the importance ofintegrating processing, microstructure, and properties in ML models."
ChangchunPeople's Republic of ChinaA siaCyborgsEmerging TechnologiesMachine LearningNanocompositesNanotechn ologyChinese Academy of Sciences
NETL
NSTL
