Thermal stability prediction of copolymerized polyimides via an interpretable transfer learning model

Journal of Materials Informatics, Год журнала: 2024, Номер 4(2)

Опубликована: Июнь 17, 2024

To address the issues with molecular representation of copolymerized polyimides (PIs) and mini dataset PI powders. We constructed an interpretable machine learning (ML) model for films using weighted-additive Morgan Fingerprints Frequency descriptors developed transfer enhance Thermal Stability (Temperature at 5% weight loss) powders, it is recommended to add conjugated functional groups diamines, control phenyl ring side chains, reduce pyridine hydroxyl groups; select copolyimides (co-PIs); ensure that anhydride directly connected benzene in dianhydrides, avoiding aliphatic cycles. It noteworthy close alignment between experimental results predictions serves confirm a reliable prediction tool. hoped this polymer informatics approach will provide further implementation practical applications other materials.

Язык: Английский

Machine learning for thermal transport

Journal of Applied Physics, Год журнала: 2024, Номер 136(16)

Опубликована: Окт. 24, 2024

Язык: Английский

Harnessing quantum power: Revolutionizing materials design through advanced quantum computation

Materials Genome Engineering Advances, Год журнала: 2024, Номер unknown

Опубликована: Дек. 4, 2024

Abstract The design of advanced materials for applications in areas photovoltaics, energy storage, and structural engineering has made significant strides. However, the rapid proliferation candidate materials—characterized by complexity that complicates relationships between features—presents substantial challenges manufacturing, fabrication, characterization. This review introduces a comprehensive methodology using cutting‐edge quantum computing, with particular focus on quadratic unconstrained binary optimization (QUBO) machine learning (QML). We introduce loop framework QUBO‐empowered design, including constructing high‐quality datasets capture critical material properties, employing tailored computational methods precise modeling, developing figures merit to evaluate performance metrics, utilizing algorithms discover optimal materials. In addition, we delve into core principles QML illustrate its transformative potential accelerating discovery through range simulations innovative adaptations. also highlights active strategies integrate artificial intelligence, offering more efficient pathway explore vast, complex space. Finally, discuss key future opportunities emphasizing their revolutionize field facilitate groundbreaking innovations.

Язык: Английский