ANN-based prediction of torsional fatigue behavior in an injection-molded compliant mechanism component
- 1Faculty of Advanced Education, Ho Chi Minh City University of Technology and Engineering, Ho Chi Minh City, Vietnam.
- 2Faculty of Mechanical Engineering, Ho Chi Minh City University of Technology and Engineering, Ho Chi Minh City, Vietnam.
Abstract
An Artificial Neural Network (ANN) model was created to predict how injection-molded parts made from flexible mechanisms would hold up under twisting fatigue. These parts were made from three common plastics—ABS (Acrylonitrile Butadiene Styrene), PP (polypropylene), and PA6 (Polyamide 6)—strengthened with up to 30 wt.% short glass fibers. The team also looked at how many times the parts were cycled during manufacturing.To test them, the samples were twisted back and forth ±40° for 100 cycles. The study examined how the type of plastic, the amount of glass fiber, and key injection molding settings affected the results. It turned out that PA6’s ability to withstand fatigue really improved when it had more glass fiber, especially 20 wt.% or more. PA6 turned out to be the strongest material in this test, followed by ABS. PP was the most consistent but also the weakest in terms of twisting strength. The ANN model was quite successful at predicting the twisting force, with a match range (R ≈ 0.94 to 0.98) and cross validation 0.93 and 0.97. The research also pointed out specific injection molding conditions—like melt temperature, filling pressure, and packing pressure—as being particularly important for the final performance. Overall, this proposed approach offers a tool to help engineers design and make plastic flexible mechanisms more effectively.
Related
- Mechanism-based prediction of insertion-driven high pathogenicity avian influenza virus emergence
- MRFlow: Mechanism-Structured Residual Flow Matching for Virtual Cell Perturbation Prediction
- 3D Human Neuromuscular Architecture Encoded by Compliant Patterned Hydrogel Scaffolds
- Beyond Ramachandran distributions: quantifying secondary structure effects on amino acids and their pairs in the full torsional space
- Design and validation of a GMP-compliant manufacturing platform of extracellular vesicles from mesenchymal stromal cells
BibTeX
@article{huy2026annbased,
title = {{ANN-based prediction of torsional fatigue behavior in an injection-molded compliant mechanism component}},
author = {Vu Quang Huy and Tran Minh The Uyen and Pham Son Minh and Van-Thuc Nguyen and Hai Nguyen Le Dang},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0361201},
url = {https://doi.org/10.1371/journal.pone.0361201},
note = {CC BY},
}
RIS
TY - JOUR
TI - ANN-based prediction of torsional fatigue behavior in an injection-molded compliant mechanism component
AU - Vu Quang Huy
AU - Tran Minh The Uyen
AU - Pham Son Minh
AU - Van-Thuc Nguyen
AU - Hai Nguyen Le Dang
PY - 2026
DA - 2026/10/28
JO - PLOS One
DO - 10.1371/journal.pone.0361201
UR - https://doi.org/10.1371/journal.pone.0361201
AB - An Artificial Neural Network (ANN) model was created to predict how injection-molded parts made from flexible mechanisms would hold up under twisting fatigue. These parts were made from three common plastics—ABS (Acrylonitrile Butadiene Styrene), PP (polypropylene), and PA6 (Polyamide 6)—strengthened with up to 30 wt.% short glass fibers. The team also looked at how many times the parts were cycled during manufacturing.To test them, the samples were twisted back and forth ±40° for 100 cycles. The study examined how the type of plastic, the amount of glass fiber, and key injection molding settings affected the results. It turned out that PA6’s ability to withstand fatigue really improved when it had more glass fiber, especially 20 wt.% or more. PA6 turned out to be the strongest material in this test, followed by ABS. PP was the most consistent but also the weakest in terms of twisting strength. The ANN model was quite successful at predicting the twisting force, with a match range (R ≈ 0.94 to 0.98) and cross validation 0.93 and 0.97. The research also pointed out specific injection molding conditions—like melt temperature, filling pressure, and packing pressure—as being particularly important for the final performance. Overall, this proposed approach offers a tool to help engineers design and make plastic flexible mechanisms more effectively.
SN - 1932-6203
ER -