Machine learning (ANFIS) optimization of h-BN and GNP-based polymer composite.
In: AIP Conference Proceedings, Jg. 3122 (2024-05-29), Heft 1, S. 1-7
Konferenz
Zugriff:
The utilization of advanced materials in polymer composites has gained significant attention due to their enhanced properties and performance in various applications. The current work is an attempt to optimize the mechanical properties of GNP (graphene nanoplatelets) and h-BN (Hexagonal Boron Nitride) based polymer composites using fuzzy and neural-based Adaptive Neuro-Fuzzy Inference System (ANFIS). The current work investigates the effects of nanofiller (both GNP & h-BN) content on the composite's mechanical properties (tensile and flexure strength). A series of tensile and flexure experiments were conducted to fabricate composite samples with varying concentrations of GNP and h-BN. The ANFIS algorithm was employed as a predictive modeling tool to optimize the composite properties. The ANFIS model was trained and tested using the data gathered from the experimental trials, and its ability to predict the mechanical properties of the composite was assessed. Various input parameters, including GNP, h-BN content, and dosage, were considered to determine the optimal composition for maximizing the desired mechanical properties. The training results of membership function Trimf under backpropagation and Dsigmf under hybrid configuration have shown the least root mean square error (RMSE) of 0.022842 and 3.8043 for Tensile and Flexure tests respectively. [ABSTRACT FROM AUTHOR]
Titel: |
Machine learning (ANFIS) optimization of h-BN and GNP-based polymer composite.
|
---|---|
Autor/in / Beteiligte Person: | Choukimath, Mantesh C. ; Banapurmath, N. R. |
Zeitschrift: | AIP Conference Proceedings, Jg. 3122 (2024-05-29), Heft 1, S. 1-7 |
Quelle: | 2024, Vol. 3122 Issue 1, p1-7. 7p.; Jg. 3122 (2024-05-29) 1, S. 1-7 |
Veröffentlichung: | 2024 |
Medientyp: | Konferenz |
ISSN: | 0094-243X (print) |
DOI: | 10.1063/5.0216005 |
Schlagwort: |
|
Sonstiges: |
|