A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixing
Constructing a mapping relationship among material preparation process, microstructure, and mechanical properties is a challenge in material research and development. In this work, a deep learning framework for multimodal data fusion is constructed that couples a multi-layer perceptron (MLP) and a r...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Materials & Design |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0264127524008797 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846113803957174272 |
|---|---|
| author | Zhenhua Wang Pengzhan Wang Yunfei Liu Yuanming Liu Tao Wang |
| author_facet | Zhenhua Wang Pengzhan Wang Yunfei Liu Yuanming Liu Tao Wang |
| author_sort | Zhenhua Wang |
| collection | DOAJ |
| description | Constructing a mapping relationship among material preparation process, microstructure, and mechanical properties is a challenge in material research and development. In this work, a deep learning framework for multimodal data fusion is constructed that couples a multi-layer perceptron (MLP) and a residual neural network (ResNet) to predict mechanical properties of SUS316 stainless steel ultrathin strips. Specifically, the MLP branch is used to extract the rolling process data features, and the ResNet with the addition of a convolutional block attention module (CBAM) is used to extract the microstructure features. Six models are constructed for comparison under the comprehensive consideration of factors such as unimodal network, the multimodal network and input form of image samples. The results show that the multimodal data model fused with the ResNet and MLP after adding the CBAM using both rolling process data and four types of microstructure image data as model inputs has the most accurate prediction results. The R2, MAPE, RMSE and MAE are 0.998, 0.727, 4.440 and 3.359, respectively. In addition, the proposed model is used for predicting yield strength and elongation, and the results show that the R2 values of both models on the test set are greater than 0.980, fully confirming that the multimodal data model has high prediction accuracy and good generalizability. |
| format | Article |
| id | doaj-art-9d8ef986c25d45aeb4bdc6c6e6b7ebae |
| institution | Kabale University |
| issn | 0264-1275 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Materials & Design |
| spelling | doaj-art-9d8ef986c25d45aeb4bdc6c6e6b7ebae2024-12-21T04:27:38ZengElsevierMaterials & Design0264-12752024-12-01248113504A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixingZhenhua Wang0Pengzhan Wang1Yunfei Liu2Yuanming Liu3Tao Wang4College of Mechanical and Vehicle Engineering, Taiyuan University of Technology, 030024 Taiyuan, Shanxi Province, PR China; National Key Laboratory of Metal Forming Technology and Heavy Equipment, 710018 Xi’an, Shaanxi Province, PR China; Engineering Research Center of Advanced Metal Composites Forming Technology and Equipment, Ministry of Education, 030024 Taiyuan, Shanxi Province, PR China; Corresponding author.College of Mechanical and Vehicle Engineering, Taiyuan University of Technology, 030024 Taiyuan, Shanxi Province, PR China; Engineering Research Center of Advanced Metal Composites Forming Technology and Equipment, Ministry of Education, 030024 Taiyuan, Shanxi Province, PR ChinaNational Key Laboratory of Metal Forming Technology and Heavy Equipment, 710018 Xi’an, Shaanxi Province, PR ChinaCollege of Mechanical and Vehicle Engineering, Taiyuan University of Technology, 030024 Taiyuan, Shanxi Province, PR China; National Key Laboratory of Metal Forming Technology and Heavy Equipment, 710018 Xi’an, Shaanxi Province, PR China; Engineering Research Center of Advanced Metal Composites Forming Technology and Equipment, Ministry of Education, 030024 Taiyuan, Shanxi Province, PR ChinaCollege of Mechanical and Vehicle Engineering, Taiyuan University of Technology, 030024 Taiyuan, Shanxi Province, PR China; National Key Laboratory of Metal Forming Technology and Heavy Equipment, 710018 Xi’an, Shaanxi Province, PR China; Engineering Research Center of Advanced Metal Composites Forming Technology and Equipment, Ministry of Education, 030024 Taiyuan, Shanxi Province, PR ChinaConstructing a mapping relationship among material preparation process, microstructure, and mechanical properties is a challenge in material research and development. In this work, a deep learning framework for multimodal data fusion is constructed that couples a multi-layer perceptron (MLP) and a residual neural network (ResNet) to predict mechanical properties of SUS316 stainless steel ultrathin strips. Specifically, the MLP branch is used to extract the rolling process data features, and the ResNet with the addition of a convolutional block attention module (CBAM) is used to extract the microstructure features. Six models are constructed for comparison under the comprehensive consideration of factors such as unimodal network, the multimodal network and input form of image samples. The results show that the multimodal data model fused with the ResNet and MLP after adding the CBAM using both rolling process data and four types of microstructure image data as model inputs has the most accurate prediction results. The R2, MAPE, RMSE and MAE are 0.998, 0.727, 4.440 and 3.359, respectively. In addition, the proposed model is used for predicting yield strength and elongation, and the results show that the R2 values of both models on the test set are greater than 0.980, fully confirming that the multimodal data model has high prediction accuracy and good generalizability.http://www.sciencedirect.com/science/article/pii/S0264127524008797Mechanical property predictionSUS316 stainless steel ultrathin stripsMultimodal data couplingRolling processConvolutional neural network |
| spellingShingle | Zhenhua Wang Pengzhan Wang Yunfei Liu Yuanming Liu Tao Wang A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixing Materials & Design Mechanical property prediction SUS316 stainless steel ultrathin strips Multimodal data coupling Rolling process Convolutional neural network |
| title | A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixing |
| title_full | A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixing |
| title_fullStr | A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixing |
| title_full_unstemmed | A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixing |
| title_short | A prediction model for the mechanical properties of SUS316 stainless steel ultrathin strip driven by multimodal data mixing |
| title_sort | prediction model for the mechanical properties of sus316 stainless steel ultrathin strip driven by multimodal data mixing |
| topic | Mechanical property prediction SUS316 stainless steel ultrathin strips Multimodal data coupling Rolling process Convolutional neural network |
| url | http://www.sciencedirect.com/science/article/pii/S0264127524008797 |
| work_keys_str_mv | AT zhenhuawang apredictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT pengzhanwang apredictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT yunfeiliu apredictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT yuanmingliu apredictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT taowang apredictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT zhenhuawang predictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT pengzhanwang predictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT yunfeiliu predictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT yuanmingliu predictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing AT taowang predictionmodelforthemechanicalpropertiesofsus316stainlesssteelultrathinstripdrivenbymultimodaldatamixing |