Composite Structures, Journal Year: 2024, Volume and Issue: 353, P. 118717 - 118717
Published: Nov. 16, 2024
Language: Английский
Composite Structures, Journal Year: 2024, Volume and Issue: 353, P. 118717 - 118717
Published: Nov. 16, 2024
Language: Английский
Applied Acoustics, Journal Year: 2025, Volume and Issue: 234, P. 110633 - 110633
Published: March 3, 2025
Language: Английский
Citations
2Polymer Composites, Journal Year: 2025, Volume and Issue: unknown
Published: Jan. 25, 2025
Abstract This work presents a new method to predict the transversal and shear properties of unidirectional composites (UD) through combining experimental, numerical machine learning methods. The experimental results proved complexity difficulty explaining primary factors affecting mechanical UD. representative unit cell model was then created generate 500 virtual samples for learning. show that back propagation neural network (BP) is most suitable predicting UD, with an accuracy 98% within 2% error. minimum mean square absolute errors are 1.09E‐3 1.15E‐5, respectively. It interface has significant influences on all UD modulus composite in 12 directions (G c ) affected by input parameters optimized BP model. Due wide coverage data, proposed universal can be adopted made from different kinds fibers. Highlights Interface composites. Shear along intricated. Machine Specific beneficial improve predicted accuracy.
Language: Английский
Citations
1Composite Structures, Journal Year: 2025, Volume and Issue: unknown, P. 119050 - 119050
Published: March 1, 2025
Language: Английский
Citations
1Defence Technology, Journal Year: 2025, Volume and Issue: unknown
Published: April 1, 2025
Language: Английский
Citations
1Thin-Walled Structures, Journal Year: 2024, Volume and Issue: unknown, P. 112860 - 112860
Published: Dec. 1, 2024
Language: Английский
Citations
3Published: Jan. 1, 2025
Acoustic metamaterials are artificial structures that possess distinctive acoustic characteristics, allowing for modulation effects challenging to achieve in the natural world. Nevertheless, design of is a process due intricate relationship between their structural parameters and nonlinear performance. In view limitations conventional methodologies, which rely on priori knowledge experts often hindered by prolonged computation times necessity iterative trials objectives, this paper introduces deep learning-based method performance prediction inverse Cylindrical Plate-type Metamaterials (CPAMs). The creation dataset initiated generating large number samples using parametric model, with bandgap characteristics calculated through finite element method. A forward-design learning model then developed, predicting upper lower limits based input parameters. Additionally, an constructed, enabling rapid generation desired results validated simulation experimentation, confirming accuracy reliability model. This study demonstrates potential efficiently designing complex metamaterials, offering promising solution CPAMs development.
Language: Английский
Citations
0Composite Structures, Journal Year: 2025, Volume and Issue: unknown, P. 118967 - 118967
Published: Feb. 1, 2025
Language: Английский
Citations
0Composites Part B Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 112515 - 112515
Published: April 1, 2025
Language: Английский
Citations
0Composite Structures, Journal Year: 2025, Volume and Issue: unknown, P. 119236 - 119236
Published: May 1, 2025
Language: Английский
Citations
0Published: Jan. 1, 2025
Language: Английский
Citations
0