Prediction of temperature-frequency-dependent mechanical properties of composites based on thermoplastic liquid resin reinforced with carbon fibers using artificial neural networks

A considerable interest has been generated in recent years in the use of thermoplastic polymers as matrices in the manufacture of advanced composites that require high reliability during long-term operations. In this research, a new Elium® acrylic matrix developed by Arkema was studied to evaluate t...

Full description

Saved in:
Bibliographic Details
Published in:International journal of advanced manufacturing technology 2019-12, Vol.105 (5-6), p.2543-2556
Main Authors: Barbosa, Lorena Cristina Miranda, Gomes, Guilherme, Junior, Antonio Carlos Ancelotti
Format: Article
Language:English
Subjects:
Accelerated tests
Addition polymerization
Artificial neural networks
CAE) and Design
Carbon fiber reinforced plastics
Chain mobility
Computer-Aided Engineering (CAD
Damping
Engineering
Glass transition
High frequencies
Industrial and Production Engineering
Internal friction
Life prediction
Mechanical Engineering
Mechanical properties
Media Management
Neural networks
Original Article
Polymer matrix composites
Stiffness
Superposition (mathematics)
Temperature
Temperature dependence
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A considerable interest has been generated in recent years in the use of thermoplastic polymers as matrices in the manufacture of advanced composites that require high reliability during long-term operations. In this research, a new Elium® acrylic matrix developed by Arkema was studied to evaluate the accelerated test methodology based on time-temperature superposition principle of Carbon Fiber/Elium® 150 composites. The results show that the high frequencies increase the glass transition (Tg) to higher values because the free volume is favored by polymer chains movement. In addition, artificial neural network has been used to model the temperature-frequency dependence of dynamic mechanical over the wide range of temperatures and frequencies due to its complex non-linear behavior. It has been observed that low frequencies result in low damping due to the lower internal friction, while high frequencies provide greater stiffness to the chains, resulting in a high damping. The long-term life prediction using master curves confirms that this new material can be considered to acoustic or vibrational damping purposes, considering its use in temperatures above Tg.
ISSN:0268-3768
1433-3015
DOI:10.1007/s00170-019-04486-4