Session: 04-02 Emerging Technologies: Novel AM Design

Paper Number: 98199

98199 - Machine Learning-Based Prediction of Bulk Tensile Behavior of Glass Fiber-Reinforced Polymer Composites 

There is an increasing demand for the discovery of advanced composite materials in response to the growing concern of environmental and economic issues. Therefore, screening new materials and revealing their structure-property relationships is of vital importance. The mechanical properties of traditional materials such as steel and Aluminum alloys are well-documented and predictable using theoretical methods. On the other hand, the use of composites poses two unavoidable challenges. The first is that by their nature, the use of composites introduces a wide range of compositional design space. The second is that polymers by their nature, are highly sensitive to processing, environmental, and testing conditions. Given that stress-strain curves are an important measure of a material’s mechanical properties, the current research aims to propose a machine learning approach to characterize and predict stress-strain curves of a number of glass fiber-reinforced polymer composites. Three distinct feature variables including temperature, filler content, and strain were identified to predict the output stress amounts. After data management and model development, stress-strain curves of intended composite materials are predicted and then verified by the existing experimental data. The outcomes obtained will pave the way for automated design and characterization of advanced composites without doing further experiments in smart manufacturing and Industry 4.0.

Presenting Author: Hamed Esmaeili York University

Presenting Author Biography: To live is to constantly seek improvement, and I have always had a passion for seeking new ways to innovate in daily life. Thanks to the successful completion of my bachelor’s and master’s degrees, it is glaring that studying and doing research are the endeavors I would like to engage in even more. While studying as an undergraduate student at the University of Tabriz (Iran), I found the gateway to the beautiful garden of Mechanical Engineering and developed a special interest in Manufacturing Technology. Then, I received the privilege of MSc study in Mechanical Engineering at one of the top technical universities in my country, namely Amirkabir University of Technology (Iran). In my master’s thesis, I developed an efficient machining strategy for grinding fiber-reinforced ceramic matrix composites (CMCs), which have the potential for widespread applications in modern industries due to their excellent mechanical and chemical properties. From my master’s research, I have published 5 papers in descent international journals, such as ‘Tribology International’, 'Ceramics International', 'The International Journal of Advanced Manufacturing Technology', etc. Currently, in my doctoral study, I am pursuing my research in Mechanical Engineering with a special focus on machine learning applications in the field.

Authors: