基于卷积神经网络的多孔材料有效扩散系数预测
|
摘要
提出了一种利用卷积神经网络预测多孔材料内有效扩散系数的方法,其中多孔材料微观结构的训练样本通过计算机随机模拟生成,对应的有效扩散系数通过有限元方法计算,并联用Matlab和Comsol实现,卷积神经网络训练在单块NVIDIA K80 GPU上进行,训练过程中出现的过拟合现象通过Dropout进行缓解。训练后的卷积神经网络对测试集的预测精确度达96.70%。利用这种方法,能够通过多孔材料的显微图片快速和准确计算其有效扩散系数。
In this work, a convolutional neural network was utilized to train and predict the effective diffusivity of porous material. The microstructure of porous material was generated by python script and the effective diffusivity was calculated by finite element method and using Matlab and Comsol. The training of convolutional neural network was carried on a NVIDIA K80 GPU of Google Co-laboratory, and Dropout method was utilized to reduce the over-fitting during the training process. Finally, the prediction accuracy of the trained convolutional neural network on the test samples reaches as high as 96.70%. This method can effectively and precisely predicts the effective diffusivity of porous material by using its graph information such as scanning electron microscope(SEM) pictures.
In this work, a convolutional neural network was utilized to train and predict the effective diffusivity of porous material. The microstructure of porous material was generated by python script and the effective diffusivity was calculated by finite element method and using Matlab and Comsol. The training of convolutional neural network was carried on a NVIDIA K80 GPU of Google Co-laboratory, and Dropout method was utilized to reduce the over-fitting during the training process. Finally, the prediction accuracy of the trained convolutional neural network on the test samples reaches as high as 96.70%. This method can effectively and precisely predicts the effective diffusivity of porous material by using its graph information such as scanning electron microscope(SEM) pictures.
引文
[1]Bottcher C J F.Theory of electric polarization[M].Elsevier,1952:399-422.
[2]Zimmerman R W.Thermal conductivity of fluid-saturated rocks[J].Journal of Petroleum Science and Engineering,1989,3(3):219-227.
[3]Kreher W,Pompe W.Internal stresses in heterogeneous solids[M].Akademie-Verlag,Berlin,1989:1-225.
[4]Wong C P,Bollampally R S.Thermal conductivity,elastic modulus,and coefficient of thermal expansion of polymer composites filled with ceramic particles for electronic packaging[J].Journal of Applied Polymer Science,1999,74(14):3396-3403.
[5]Orrhede M,Tolani R,Salama K.Elastic constants and thermal expansion of aluminum-Si C metal-matrix composites[J].Research in Nondestructive Evaluation,1996,8(1):23-37.
[6]Landauer R.The electrical resistance of binary metallic mixtures[J].Journal of Applied Physics,1952,23(7):779-784.
[7]Bruggeman D A G.Berechnung verschiedener physikalischer Konstanten von heterogenen Substanzen.I.Dielektrizit?tskonstanten und Leitf?higkeiten der Mischk?rper aus isotropen Substanzen[J].Annalen Der Physik,1935,(24):636-664.
[8]Bruggeman D A G.Berechnung verschiedener physikalischer Konstanten von heterogenen Substanzen.II.Dielektrizit?tskonstanten und Leitf?higkeiten von Vielkristallen der nichtregul?ren Systeme[J].Annalen Der Physik,1935,24:665-679.
[9]Maxwell J C.A treatise on Electricity and Magnetism[M].Oxford University Press,Cambridge,UK,1904:360-373.
[10]Hamilton R L,Crosser O K.An experimental determination of the thermal conductivity of several greases[J].Industrial&Engineering Chemistry Fundamentals,1962,7(1):187.
[11]Gao L,Gu J Z.Effective dielectric constant of a two-component material with shape distribution[J].Journal of Physics D:Applied Physics,2002,35(3):267-271.
[12]Wang M,Pan N.Predictions of effective physical properties of complex multiphase materials[J].Materials Science and Engineering R,2008,63(1):1-30.
[13]Rio J A,Zimmerman R W,Dawe R A.Formula for the conductivity of a two-component material based on the reciprocity theorem[J].Solid State Communications,1998,106(4):183-186.
[14]Le Cun Y,Bengio Y,Hinton G.Deep learning[J].Nature,2015,521:436-444.
[15]周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报,2017,40(6):1229-1251.Zhou Feiyan,Jin Linpeng,Dong Jun.Review of convolutional neural network[J].Chinese Journal of Computers,2017,40(6):1229-1251.
[16]Goodfellow I,Bengio Y,Courvile A.Deep learning[M].The MIT Press,2016:274-305.
[17]Chollet F.Deep learning with python[M].Manning Publications Co,2018:1-24.
[18]Calame J P.Finite difference simulations of permittivity and electric field statistics in ceramic-polymer composites for capacitor applications[J].Journal of Applied Physics,2006,99(8):706-269.
[19]Simonyan K,Zisserman A.Very deep convolutional networks for large-scale image recognition[C]//In ICLR,2015:1-14.
[20]He K M,Zhang X Y,Ren S Q,et al.Deep residual learning for image recognition[C]//In CVPR,2016:770-778.
[21]Szegedy C,Ioffe S,Vanhoucke V,et al.Inception-v4,Inception-Res Net and the impact of residual connections on learning[C]//Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence,2017:4278-4284.
[22]Srivastava N,Hinton G,Krizhevsky A,et al.Dropout:A simple way to prevent neural networks from overfitting[J].Journal of Machine Learning Research,2014,15:1929-1958.
[2]Zimmerman R W.Thermal conductivity of fluid-saturated rocks[J].Journal of Petroleum Science and Engineering,1989,3(3):219-227.
[3]Kreher W,Pompe W.Internal stresses in heterogeneous solids[M].Akademie-Verlag,Berlin,1989:1-225.
[4]Wong C P,Bollampally R S.Thermal conductivity,elastic modulus,and coefficient of thermal expansion of polymer composites filled with ceramic particles for electronic packaging[J].Journal of Applied Polymer Science,1999,74(14):3396-3403.
[5]Orrhede M,Tolani R,Salama K.Elastic constants and thermal expansion of aluminum-Si C metal-matrix composites[J].Research in Nondestructive Evaluation,1996,8(1):23-37.
[6]Landauer R.The electrical resistance of binary metallic mixtures[J].Journal of Applied Physics,1952,23(7):779-784.
[7]Bruggeman D A G.Berechnung verschiedener physikalischer Konstanten von heterogenen Substanzen.I.Dielektrizit?tskonstanten und Leitf?higkeiten der Mischk?rper aus isotropen Substanzen[J].Annalen Der Physik,1935,(24):636-664.
[8]Bruggeman D A G.Berechnung verschiedener physikalischer Konstanten von heterogenen Substanzen.II.Dielektrizit?tskonstanten und Leitf?higkeiten von Vielkristallen der nichtregul?ren Systeme[J].Annalen Der Physik,1935,24:665-679.
[9]Maxwell J C.A treatise on Electricity and Magnetism[M].Oxford University Press,Cambridge,UK,1904:360-373.
[10]Hamilton R L,Crosser O K.An experimental determination of the thermal conductivity of several greases[J].Industrial&Engineering Chemistry Fundamentals,1962,7(1):187.
[11]Gao L,Gu J Z.Effective dielectric constant of a two-component material with shape distribution[J].Journal of Physics D:Applied Physics,2002,35(3):267-271.
[12]Wang M,Pan N.Predictions of effective physical properties of complex multiphase materials[J].Materials Science and Engineering R,2008,63(1):1-30.
[13]Rio J A,Zimmerman R W,Dawe R A.Formula for the conductivity of a two-component material based on the reciprocity theorem[J].Solid State Communications,1998,106(4):183-186.
[14]Le Cun Y,Bengio Y,Hinton G.Deep learning[J].Nature,2015,521:436-444.
[15]周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报,2017,40(6):1229-1251.Zhou Feiyan,Jin Linpeng,Dong Jun.Review of convolutional neural network[J].Chinese Journal of Computers,2017,40(6):1229-1251.
[16]Goodfellow I,Bengio Y,Courvile A.Deep learning[M].The MIT Press,2016:274-305.
[17]Chollet F.Deep learning with python[M].Manning Publications Co,2018:1-24.
[18]Calame J P.Finite difference simulations of permittivity and electric field statistics in ceramic-polymer composites for capacitor applications[J].Journal of Applied Physics,2006,99(8):706-269.
[19]Simonyan K,Zisserman A.Very deep convolutional networks for large-scale image recognition[C]//In ICLR,2015:1-14.
[20]He K M,Zhang X Y,Ren S Q,et al.Deep residual learning for image recognition[C]//In CVPR,2016:770-778.
[21]Szegedy C,Ioffe S,Vanhoucke V,et al.Inception-v4,Inception-Res Net and the impact of residual connections on learning[C]//Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence,2017:4278-4284.
[22]Srivastava N,Hinton G,Krizhevsky A,et al.Dropout:A simple way to prevent neural networks from overfitting[J].Journal of Machine Learning Research,2014,15:1929-1958.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |