7th IMRMPT

Abstract

Abstract information

English title:	Fatigue behavior prediction of additive manufacturing composites reinforced with long fibers using machine learning
Spanish title:	Predicci�n del comportamiento a fatiga de materiales compuestos de manufactura aditiva reforzados con fibras largas usando machine learning
English abstract:	Additive manufacturing is a versatile manufacturing process that is quite useful for complex parts for which traditional methods would be difficult or outright impossible to fabricate. However, the technique has limitations associated with the low strength and stiffness characteristics of thermoplastic materials. For this reason, techniques such as Fused Deposition Modeling (FDM) have been discarded during the design process when functional load-bearing components are desired. In this research, experimental data from mechanical tests, tensile strength, and fatigue are analyzed to characterize the mechanical properties of Continuous Fiber Reinforced Thermoplastic Composites (CFRTPC). Nylon is used for the matrix filler, with triangular or hexagonal patterns. For reinforcement, fiberglass, carbon fiber, and Kevlar are used, using concentric rings and reinforcement layers with fiber in different orientations. The data processing was done through Machine Learning (ML), and regressions were obtained to characterize the behavior of the different compositions under the test stresses. The results showed that machine learning models were able to accurately predict the fatigue behavior of the composites with reasonable accuracy. Moreover, these models have the potential to be used for designing and optimizing composites produced by additive manufacturing processes.
Spanish abstract:	La manufactura aditiva es un proceso de fabricaci�n vers�til y �til para piezas complejas, para las cuales usar m�todos tradicionales ser�a dif�cil o directamente imposible de fabricar. Sin embargo, la t�cnica presenta limitaciones asociadas a la baja resistencia y rigidez caracter�sticas de los materiales termopl�sticos. Por esta raz�n, t�cnicas como el modelado por deposici�n fundida (FDM) se han descartado durante el proceso de dise�o, cuando se desea que dicho componente resista cargas funcionales. En esta investigaci�n se analizan datos experimentales de ensayos mec�nicos, de resistencia a tracci�n y fatiga, que permiten caracterizar las propiedades mec�nicas de compuestos termopl�stico reforzados con fibras continuas (CFRTPC). Para la matriz se utiliza nylon y patrones de relleno triangulares o hexagonales. Para el refuerzo se utiliza fibra de vidrio, fibra de carbono y Kevlar, usando anillos conc�ntricos y capas de refuerzo con fibra en diferentes orientaciones. El procesamiento de los datos obtenidos se hace utilizando Machine Learning (ML), desarrollando regresiones para caracterizar el comportamiento mec�nico de los compuestos bajo diferentes par�metros de proceso. Los resultados muestran que los modelos de aprendizaje autom�tico son capaces de predecir el comportamiento a fatiga de los materiales compuestos con una precisi�n razonable. Adem�s, estos modelos tienen el potencial de ser utilizados para dise�ar y optimizar materiales compuestos producidos por procesos de fabricaci�n aditiva.
Topic:
Presentation Type:
Speaker:

Authors

Name	Institution	Email
Hern�n David S�nchez Restrepo	Universidad Industrial de Santander	hernan2190363@correo.uis.edu.co
Cristian Andr�s Hern�ndez Salazar	Universidad Industrial de Santander	cristian2228339@correo.uis.edu.co
Octavio Andr�s Gonz�lez-Estrada	Universidad Industrial de Santander	agonzale@uis.edu.co

Presentation

Room:	Principal Room
Date:	June 7, 2023
Hour:	09:20 - 09:40

Back

7^th IMRMPT
Seventh International Meeting for Researchers in Materials and Plasma Technology

Previous editions:

        1st IMRMPT 2011
        2nd IMRMPT 2013
        3rd IMRMPT 2015
        4th IMRMPT 2017
        5th IMRMPT 2019
        5+1 IMRMPT 2021

Contact information:
imrmpt@foristom.org

Copyright 2022 FORISTOM
All rights reserved