A Novel Approach to Predict Green Density by HVC Based on Materials Informatics Method
High velocity compaction(HVC)is an advanced compaction technique to obtain high density compacts with a compaction velocity of ≤ 10m/s.It was applied to various kinds of metallic powders and had been verified by achieving a density level of greater than 7.5g/cm3 for the iron-based powders.Making predictions of the green density rapidly and accurately is of much practical importance,especially for costly and time-consuming material design by trial and error.In this paper,we provide a machine learning approach to predict green density using relevant material descriptors including chemical composition,powder properties and compaction energy.We try four models with the experimental dataset for appropriate model selecting,and multilayer perceptron model works well for its distinguished prediction performance with high correlation coefficient and low error values.Then applying this model,nine kinds of materials are predicted on their green density with specific processing parameters.The predicted green density agreed very well with the experimental one for each material,with the inaccuracy less than 2%.So the method is confirmed by experimental technique with respect to prediction accuracy.
Powder metallurgy High velocity compaction Green density Data mining Multilayer perceptron
Zhang Kaiqi Yin Haiqing Jiang Xue Deng Zhenghua Khan Dil Faraz Zheng Qingjun Qu Xuanhui
Collaborative Innovation Center of Steel Technology,University of Science and Technology Beijing,Bei Collaborative Innovation Center of Steel Technology,University of Science and Technology Beijing,Bei Collaborative Innovation Center of Steel Technology,University of Science and Technology Beijing,Bei Department of Physics,University of Science and Technology Bannu,Bannu 28100,Pakistan Kennametal Inc,1600 Technology Way,PA 15650,USA
国际会议
北京
英文
2056-2062
2018-09-16（万方平台首次上网日期，不代表论文的发表时间）