Multi-objective optimization of the epoxy matrix system using machine learning
The material properties of the epoxy matrix system are optimized by applying machine learning to a dataset composed of the data on the epoxy composition with the additives and the experimental material properties with missing values before and after curing. First, we construct regression models to p...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-03-01
|
| Series: | Results in Materials |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590048X23000146 |
| _version_ | 1797884513954562048 |
|---|---|
| author | Shigeru Taniguchi Kaori Uemura Shogo Tamaki Keiichiro Nomura Kohei Koyanagi Shigeru Kuchii |
| author_facet | Shigeru Taniguchi Kaori Uemura Shogo Tamaki Keiichiro Nomura Kohei Koyanagi Shigeru Kuchii |
| author_sort | Shigeru Taniguchi |
| collection | DOAJ |
| description | The material properties of the epoxy matrix system are optimized by applying machine learning to a dataset composed of the data on the epoxy composition with the additives and the experimental material properties with missing values before and after curing. First, we construct regression models to predict material properties from the information about the composition of 22 kinds of epoxies, 11 kinds of reactive agents, 8 kinds of alcohols, 15 kinds of curing agents, and 3 kinds of other additives. As machine-learning models, Partial Least Squares Regression, Support Vector Regression, Random Forest Regression, Kernel ridge regression, and Artificial Neural Networks are used. Secondly, desirable compositions are identified by applying the constructed models to many candidates of possible compositions with paying attention to the restrictions of the range of the mixing ratio. Finally, we succeed in making desirable epoxy matrix systems by adopting the identified compositions based on machine-learning predictions, and the usefulness of our approach is clearly shown. Only a few additional experiments allow us to make heat-resistant epoxy matrix systems with high processability and productivity, which have never been achieved for more than 300 experiments with human-choice compositions in a trial-error approach. |
| first_indexed | 2024-04-10T04:07:46Z |
| format | Article |
| id | doaj.art-e72e1e7b12114944a4fb420390fe299e |
| institution | Directory Open Access Journal |
| issn | 2590-048X |
| language | English |
| last_indexed | 2024-04-10T04:07:46Z |
| publishDate | 2023-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Materials |
| spelling | doaj.art-e72e1e7b12114944a4fb420390fe299e2023-03-13T04:16:06ZengElsevierResults in Materials2590-048X2023-03-0117100376Multi-objective optimization of the epoxy matrix system using machine learningShigeru Taniguchi0Kaori Uemura1Shogo Tamaki2Keiichiro Nomura3Kohei Koyanagi4Shigeru Kuchii5Department of Creative Engineering, National Institute of Technology, Kitakyushu College, Kitakyushu, 802-0985, Japan; Corresponding author.Department of Creative Engineering, National Institute of Technology, Kitakyushu College, Kitakyushu, 802-0985, JapanDepartment of Creative Engineering, National Institute of Technology, Kitakyushu College, Kitakyushu, 802-0985, JapanChemicals Research Laboratories, Toray Industries, Nagoya, 455-8502, JapanChemicals Research Laboratories, Toray Industries, Nagoya, 455-8502, JapanDepartment of Creative Engineering, National Institute of Technology, Kitakyushu College, Kitakyushu, 802-0985, JapanThe material properties of the epoxy matrix system are optimized by applying machine learning to a dataset composed of the data on the epoxy composition with the additives and the experimental material properties with missing values before and after curing. First, we construct regression models to predict material properties from the information about the composition of 22 kinds of epoxies, 11 kinds of reactive agents, 8 kinds of alcohols, 15 kinds of curing agents, and 3 kinds of other additives. As machine-learning models, Partial Least Squares Regression, Support Vector Regression, Random Forest Regression, Kernel ridge regression, and Artificial Neural Networks are used. Secondly, desirable compositions are identified by applying the constructed models to many candidates of possible compositions with paying attention to the restrictions of the range of the mixing ratio. Finally, we succeed in making desirable epoxy matrix systems by adopting the identified compositions based on machine-learning predictions, and the usefulness of our approach is clearly shown. Only a few additional experiments allow us to make heat-resistant epoxy matrix systems with high processability and productivity, which have never been achieved for more than 300 experiments with human-choice compositions in a trial-error approach.http://www.sciencedirect.com/science/article/pii/S2590048X23000146Materials informaticsEpoxy mixtureMulti-objective optimizationInverse analysis |
| spellingShingle | Shigeru Taniguchi Kaori Uemura Shogo Tamaki Keiichiro Nomura Kohei Koyanagi Shigeru Kuchii Multi-objective optimization of the epoxy matrix system using machine learning Results in Materials Materials informatics Epoxy mixture Multi-objective optimization Inverse analysis |
| title | Multi-objective optimization of the epoxy matrix system using machine learning |
| title_full | Multi-objective optimization of the epoxy matrix system using machine learning |
| title_fullStr | Multi-objective optimization of the epoxy matrix system using machine learning |
| title_full_unstemmed | Multi-objective optimization of the epoxy matrix system using machine learning |
| title_short | Multi-objective optimization of the epoxy matrix system using machine learning |
| title_sort | multi objective optimization of the epoxy matrix system using machine learning |
| topic | Materials informatics Epoxy mixture Multi-objective optimization Inverse analysis |
| url | http://www.sciencedirect.com/science/article/pii/S2590048X23000146 |
| work_keys_str_mv | AT shigerutaniguchi multiobjectiveoptimizationoftheepoxymatrixsystemusingmachinelearning AT kaoriuemura multiobjectiveoptimizationoftheepoxymatrixsystemusingmachinelearning AT shogotamaki multiobjectiveoptimizationoftheepoxymatrixsystemusingmachinelearning AT keiichironomura multiobjectiveoptimizationoftheepoxymatrixsystemusingmachinelearning AT koheikoyanagi multiobjectiveoptimizationoftheepoxymatrixsystemusingmachinelearning AT shigerukuchii multiobjectiveoptimizationoftheepoxymatrixsystemusingmachinelearning |