A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber
Recently, deep learning methods have become one of the hottest topics in predicting material properties, however, one bottleneck in current research is the simultaneous analysis of heterogeneous data. In this study, a deep learning fusion model is developed for the first time to predict the material...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-03-01
|
| Series: | Materials & Design |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0264127523001752 |
| _version_ | 1797853936536780800 |
|---|---|
| author | Mengze Li Shuran Li Yu Tian Yihan Fu Yanliang Pei Weidong Zhu Yinglin Ke |
| author_facet | Mengze Li Shuran Li Yu Tian Yihan Fu Yanliang Pei Weidong Zhu Yinglin Ke |
| author_sort | Mengze Li |
| collection | DOAJ |
| description | Recently, deep learning methods have become one of the hottest topics in predicting material properties, however, one bottleneck in current research is the simultaneous analysis of heterogeneous data. In this study, a deep learning fusion model is developed for the first time to predict the material properties of carbon fiber monofilament using textual (macroscopic properties of composites and matrix) and visual (two-point statistics of microstructures) data. For this, 1200 stochastic microstructures are generated using the greedy-based generation (GBG) algorithm. Then, the statistical representations of microstructures are determined using two-point statistics and the macroscopic properties are calculated based on a micro-scale finite element (FE) simulation. Finally, the visual and textual data are fed into the convolutional neural network (CNN) and multi-layer perceptron (MLP) fusion model for predicting the mechanical properties of carbon fibers. The developed hybrid CNN-MLP fusion model achieves encouraging average testing R2 of longitudinal modulus, transverse modulus, in-plane shear modulus, major Poisson’s ratio, and out-of-plane shear modulus of carbon fibers with values of 0.991, 0.969, 0.984, 0.903, and 0.955, respectively. Thus, the proposed strategy provides a promising framework for predicting material properties via multisource heterogeneous data and is expected to accelerate the smart design and optimization of materials. |
| first_indexed | 2024-04-09T19:58:52Z |
| format | Article |
| id | doaj.art-abcd71c29b464ff88d4132d85efe39d5 |
| institution | Directory Open Access Journal |
| issn | 0264-1275 |
| language | English |
| last_indexed | 2024-04-09T19:58:52Z |
| publishDate | 2023-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Materials & Design |
| spelling | doaj.art-abcd71c29b464ff88d4132d85efe39d52023-04-03T05:20:58ZengElsevierMaterials & Design0264-12752023-03-01227111760A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiberMengze Li0Shuran Li1Yu Tian2Yihan Fu3Yanliang Pei4Weidong Zhu5Yinglin Ke6State Key Laboratory of Fluid Power and Mechatronic System, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China; Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaState Key Laboratory of Fluid Power and Mechatronic System, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China; Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China; Corresponding author at: State Key Laboratory of Fluid Power and Mechatronic System, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China.State Key Laboratory of Fluid Power and Mechatronic System, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China; Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaState Key Laboratory of Fluid Power and Mechatronic System, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China; Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaLaboratory for Marine Geology, Laoshan Laboratory, Qingdao, China; Key Laboratory of Marine Geology and metallogeny, First Institute of Oceanography, MNR, Qingdao, ChinaState Key Laboratory of Fluid Power and Mechatronic System, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China; Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaState Key Laboratory of Fluid Power and Mechatronic System, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, China; Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaRecently, deep learning methods have become one of the hottest topics in predicting material properties, however, one bottleneck in current research is the simultaneous analysis of heterogeneous data. In this study, a deep learning fusion model is developed for the first time to predict the material properties of carbon fiber monofilament using textual (macroscopic properties of composites and matrix) and visual (two-point statistics of microstructures) data. For this, 1200 stochastic microstructures are generated using the greedy-based generation (GBG) algorithm. Then, the statistical representations of microstructures are determined using two-point statistics and the macroscopic properties are calculated based on a micro-scale finite element (FE) simulation. Finally, the visual and textual data are fed into the convolutional neural network (CNN) and multi-layer perceptron (MLP) fusion model for predicting the mechanical properties of carbon fibers. The developed hybrid CNN-MLP fusion model achieves encouraging average testing R2 of longitudinal modulus, transverse modulus, in-plane shear modulus, major Poisson’s ratio, and out-of-plane shear modulus of carbon fibers with values of 0.991, 0.969, 0.984, 0.903, and 0.955, respectively. Thus, the proposed strategy provides a promising framework for predicting material properties via multisource heterogeneous data and is expected to accelerate the smart design and optimization of materials.http://www.sciencedirect.com/science/article/pii/S0264127523001752Carbon fibersPolymer-matrix composites (PMCs)Mechanical propertiesDeep learningMultimodal fusion |
| spellingShingle | Mengze Li Shuran Li Yu Tian Yihan Fu Yanliang Pei Weidong Zhu Yinglin Ke A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber Materials & Design Carbon fibers Polymer-matrix composites (PMCs) Mechanical properties Deep learning Multimodal fusion |
| title | A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber |
| title_full | A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber |
| title_fullStr | A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber |
| title_full_unstemmed | A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber |
| title_short | A deep learning convolutional neural network and multi-layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber |
| title_sort | deep learning convolutional neural network and multi layer perceptron hybrid fusion model for predicting the mechanical properties of carbon fiber |
| topic | Carbon fibers Polymer-matrix composites (PMCs) Mechanical properties Deep learning Multimodal fusion |
| url | http://www.sciencedirect.com/science/article/pii/S0264127523001752 |
| work_keys_str_mv | AT mengzeli adeeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT shuranli adeeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yutian adeeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yihanfu adeeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yanliangpei adeeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT weidongzhu adeeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yinglinke adeeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT mengzeli deeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT shuranli deeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yutian deeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yihanfu deeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yanliangpei deeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT weidongzhu deeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber AT yinglinke deeplearningconvolutionalneuralnetworkandmultilayerperceptronhybridfusionmodelforpredictingthemechanicalpropertiesofcarbonfiber |