Related references
Note: Only part of the references are listed.
Article
Materials Science, Composites
Fengqing Chen et al.
Summary: Extensive machine learning methods have brought about significant changes in various fields such as metals, catalysts, and polymers. However, the application of machine learning in the exploration of functional polymer-based composites, particularly in flame retardancy, is still in its early stages. In this study, an adaptive framework combining domain knowledge and machine learning was proposed to accelerate the optimization of high flame retardant composites. Different data resources, including experiments, handbooks, and published papers, were used for training, feedback, or prediction purposes. The framework demonstrated an effective approach for feature engineering and classification of flame-retardant polymer-based composites. Four machine learning methods were compared in the framework, and the combination of Lasso, Ridge, and ANN showed higher accuracy in predicting the limit oxygen index (LOI), assisting in the discovery of new experiments and effective prediction of different flame retardants. The optimized models from the adaptive framework could contribute to machine intelligence in the engineering of flame-retardant polymer-based composites. Moreover, the proposed adaptive framework has the potential to be extended to the machine intelligence design of other functional polymer-based composites.
COMPOSITES SCIENCE AND TECHNOLOGY
(2023)
Article
Engineering, Environmental
Zhongwei Chen et al.
Summary: By using machine learning, a high-performance EP composite with enhanced fire resistance was developed by incorporating OPFRs. The ML model identified fire retardants with specific molecular structures that significantly increased the LOI of EPs. Experimental validation confirmed the accuracy and reliability of the ML model.
CHEMICAL ENGINEERING JOURNAL
(2023)
Review
Automation & Control Systems
Cheng Yan et al.
Summary: With the rapid development of computing power and algorithms, machine learning has shown its great potential in new polymer discovery. This article provides a history of machine learning and summarizes the basic process of machine learning accelerated polymer discovery. It also reviews the four steps involved in this process, which include dataset selection, fingerprinting, machine learning framework, and new polymer generation. The challenges and future prospects in machine learning accelerated polymer discovery are discussed.
ADVANCED INTELLIGENT SYSTEMS
(2023)
Review
Chemistry, Multidisciplinary
Bo-Wen Liu et al.
Summary: Most organic polymeric materials are highly flammable, causing significant damages to human life and property through the large amounts of smoke, toxic gases, heat, and melt drips produced during burning. Conventional flame-retardant methods are facing difficulties in meeting the increasing flame-retardant requirements. Advanced flame-retardant methods, such as all-in-one intumescence and nanotechnology, have been developed to provide potential solutions to these challenges.
ADVANCED MATERIALS
(2022)
Review
Polymer Science
Morgan M. Cencer et al.
Summary: Polymers are versatile materials and researchers are using data science and polymer informatics to design new materials and understand their structure-property relationships. Despite many useful tools and databases available, they are not widely utilized.
POLYMER INTERNATIONAL
(2022)
Review
Chemistry, Applied
Merve S. Ozer et al.
Summary: This review summarizes recent reports on the preparation of phosphorus-based flame retardant compounds and their application on various textiles through coating. It highlights the synergistic interaction of phosphorus with other environmentally acceptable flame retardant elements to improve flame retardant efficiency.
PROGRESS IN ORGANIC COATINGS
(2022)
Article
Chemistry, Multidisciplinary
Matteo Aldeghi et al.
Summary: This study introduces a graph representation and graph neural network architecture for polymer property prediction. By constructing a large dataset of polymer materials and developing corresponding machine learning models, it achieves accurate capture of critical features and outperforms traditional cheminformatics methodologies.
Article
Nanoscience & Nanotechnology
Cheng Yan et al.
Summary: This study successfully discovered five new types of UV-curable thermoset shape memory polymers using machine learning techniques. By combining transfer learning-variational autoencoder and a weighted-vector combination, researchers overcame the challenge of limited data, opening new opportunities in the field of TSMP. Through representation of TSMP features with drug molecules and development of a ML framework, the approach showed more accurate and robust results compared to traditional methods.
ACS APPLIED MATERIALS & INTERFACES
(2021)
Article
Materials Science, Composites
Hoang T. Nguyen et al.
Summary: In this study, an artificial neural network (ANN) model was developed to predict the heat release properties of composites. The comparison between different machine learning algorithms revealed that the BRANNGP model can accurately predict the heat release rate and total heat release of composites. Moreover, the BRANNGP model can eliminate outliers and accurately estimate the complex nonlinear relationship between heat release rate and heat flux exposure time.
COMPOSITES SCIENCE AND TECHNOLOGY
(2021)
Article
Polymer Science
Cheng Yan et al.
Summary: This study proposes methodologies to address the difficulties in applying ML in the field of thermoset shape memory polymers (TSMPs) and develops a new ML framework for predicting the recovery stresses of TSMPs. By using these methods, 14 mostly unknown TSMPs with greater recovery stress than the known ones were identified and one of them was validated by molecular dynamics simulation.
Editorial Material
Nanoscience & Nanotechnology
Adam J. Gormley et al.
Summary: The design of new functional polymers relies on exploring their structure-function landscapes. Advances in combinatorial polymer chemistry and machine learning offer exciting opportunities for engineering purpose-fit polymeric materials.
NATURE REVIEWS MATERIALS
(2021)
Review
Materials Science, Multidisciplinary
Lihua Chen et al.
Summary: Artificial intelligence is making significant impact in the field of polymer informatics, with surrogate models being trained on polymer data for instant property prediction. Challenges include lack of curated data availability and the need for machine-readable representations capturing the complexity of polymer structures.
MATERIALS SCIENCE & ENGINEERING R-REPORTS
(2021)
Article
Nanoscience & Nanotechnology
Fengqing Chen et al.
Summary: The research team utilized machine learning to accelerate the development of flame-retardant polymers and explored the relationship between the limit oxygen index and components through data analysis to determine the flame-retardant mechanism and components. Additionally, by wrapping nano graphene oxide on micro zinc hydroxystannate, the flame retardancy of polypropylene composites was successfully enhanced.
ACS APPLIED MATERIALS & INTERFACES
(2021)
Review
Materials Science, Multidisciplinary
Wuxin Sha et al.
Summary: Polymer informatics, utilizing machine learning models based on reliable data, accelerates performance prediction and process optimization of new polymers, opening up new possibilities for the development of polymer science and engineering.
Article
Polymer Science
Lemiye Atabek Savas et al.
Review
Nanoscience & Nanotechnology
Simone T. Lazar et al.
NATURE REVIEWS MATERIALS
(2020)
Article
Polymer Science
Khalifah A. Salmeia et al.
POLYMER DEGRADATION AND STABILITY
(2015)
Article
Polymer Science
Parina Patel et al.
POLYMERS FOR ADVANCED TECHNOLOGIES
(2011)
Article
Chemistry, Medicinal
David Rogers et al.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2010)
Article
Engineering, Civil
SD Watt et al.
FIRE SAFETY JOURNAL
(2001)
