
Acta Materialia, Год журнала: 2024, Номер unknown, С. 120704 - 120704
Опубликована: Дек. 1, 2024
Язык: Английский
Acta Materialia, Год журнала: 2024, Номер unknown, С. 120704 - 120704
Опубликована: Дек. 1, 2024
Язык: Английский
ACS Polymers Au, Год журнала: 2023, Номер 3(3), С. 239 - 258
Опубликована: Янв. 18, 2023
In the last five years, there has been tremendous growth in machine learning and artificial intelligence as applied to polymer science. Here, we highlight unique challenges presented by polymers how field is addressing them. We focus on emerging trends with an emphasis topics that have received less attention review literature. Finally, provide outlook for field, outline important areas science discuss advances from greater material community.
Язык: Английский
Процитировано
88Journal of Hazardous Materials, Год журнала: 2023, Номер 452, С. 131243 - 131243
Опубликована: Март 21, 2023
Язык: Английский
Процитировано
25Progress in Polymer Science, Год журнала: 2024, Номер 153, С. 101828 - 101828
Опубликована: Май 3, 2024
Язык: Английский
Процитировано
14Small, Год журнала: 2024, Номер 20(28)
Опубликована: Фев. 13, 2024
Abstract High‐entropy oxide micro/nano materials (HEO MNMs) have shown broad application prospects and become hot in recent years. This review comprehensively provides an overview of the latest developments covers key aspects HEO MNMs, by discussing design principles, computer‐aided structural design, synthesis challenges strategies, as well areas. The analysis process includes role high‐throughput large‐scale HEOs along with effects temperature elevation undercooling on formation MNMs. Additionally, article summarizes high‐precision situ characterization devices field offering robust support for related research. Finally, a brief introduction to main applications MNMs is provided, emphasizing their performances. offers valuable guidance future research outlining critical issues current field.
Язык: Английский
Процитировано
9Environmental Science & Technology, Год журнала: 2024, Номер unknown
Опубликована: Дек. 16, 2024
Polymeric membranes have been widely used for liquid and gas separation in various industrial applications over the past few decades because of their exceptional versatility high tunability. Traditional trial-and-error methods material synthesis are inadequate to meet growing demands high-performance membranes. Machine learning (ML) has demonstrated huge potential accelerate design discovery membrane materials. In this review, we cover strengths weaknesses traditional methods, followed by a discussion on emergence ML developing advanced polymeric We describe methodologies data collection, preparation, commonly models, explainable artificial intelligence (XAI) tools implemented research. Furthermore, explain experimental computational validation steps verify results provided these models. Subsequently, showcase successful case studies emphasize inverse methodology within ML-driven structured framework. Finally, conclude highlighting recent progress, challenges, future research directions advance next generation With aim provide comprehensive guideline researchers, scientists, engineers assisting implementation process.
Язык: Английский
Процитировано
8Journal of Materials Informatics, Год журнала: 2025, Номер 5(2)
Опубликована: Март 24, 2025
Over the past decades, machine learning has kept playing an important role in materials design and discovery. In practical applications, usually need to fulfill requirements of multiple target properties. Therefore, multi-objective optimization based on become one most promising directions. This review aims provide a detailed discussion learning-assisted discovery combined with recent research progress. First, we briefly introduce workflow learning. Then, Pareto fronts corresponding algorithms are summarized. Next, strategies demonstrated, including front-based strategy, scalarization function, constraint method. Subsequently, progress is summarized different discussed. Finally, propose future directions for learning-based materials.
Язык: Английский
Процитировано
1Journal of Manufacturing Processes, Год журнала: 2025, Номер 144, С. 1 - 53
Опубликована: Апрель 15, 2025
Язык: Английский
Процитировано
1IEEE Access, Год журнала: 2024, Номер 12, С. 98319 - 98325
Опубликована: Янв. 1, 2024
Benchmark datasets normally have relatively conserved relationships and low fraction of outliers, indicated from higher determination coefficient (R2) lower Mean Absolute Error (MAE) in regression model. Here inspired by the process peeling onions, we introduced a recursive data elimination (RDE) "outliers" strategy to get benchmark dataset. Outliers are labeled using William's plot residual vs leverage (recorded as RDE_W), performance was compared with that alone RDE). The validation performed single-target multiple-target ways through predictions mechanical properties including Young's modulus, tensile strength, elongation at break for 643 polyurethane elastomers (the first time this dataset has been released), compressive strength 1030 concrete samples. In way, RDE_W achieved an 8.06% increase R2 19.87% reduction MAE RDE. way improvement approximately 3%. SVM outperformed XGB, NN, RF, Lasso DT algorithms strategy. Additional tests also validated advantages over RDE generate high-quality datasets. We released code facilitate construction high quality development new approaches better understand, explore design advanced materials.
Язык: Английский
Процитировано
6Journal of Cleaner Production, Год журнала: 2023, Номер 415, С. 137812 - 137812
Опубликована: Июнь 15, 2023
Язык: Английский
Процитировано
13Fuel, Год журнала: 2023, Номер 357, С. 129725 - 129725
Опубликована: Сен. 8, 2023
Язык: Английский
Процитировано
13