• Login
    View Item 
    •   DSpace Home
    • AUS Theses & Dissertations
    • Masters Theses
    • View Item
    •   DSpace Home
    • AUS Theses & Dissertations
    • Masters Theses
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Effect of Cooling Rate on the Mechanical Properties of Glass Fiber Polypropylene Composites

    Thumbnail
    View/ Open
    35.232-2017.03 Mohammed S. Kabbani.pdf (4.446Mb)
    Date
    2017-01
    Author
    Kabbani, Mohammed S.
    Advisor(s)
    El Kadi, Hany
    Type
    Thesis
    Metadata
    Show full item record
    Description
    A Master of Science thesis in Mechanical Engineering by Mohammed S. Kabbani entitled, "Effect of Cooling Rate on the Mechanical Properties of Glass Fiber Polypropylene Composites," submitted in January 2017. Thesis advisor is Dr. Hany El Kadi. Soft and hard copy available.
    Abstract
    Composite materials have been widely used in the recent years in almost all of the industries specially the high technology industries. Properties of the thermoplastic-based composites are affected by their processing conditions. Therefore, understanding of the behavior of these materials under different processing conditions is of most importance. Artificial Neural Networks (ANN) have recently been successfully used in the study of composite materials. This study aims to predict the mechanical properties of unidirectional glassfiber polypropylene composite materials processed under different cooling rates as a function of the fiber orientation angle using ANN. Composite specimens with five different fiber orientation angles were manufactured under different cooling rates using a compression molding press. These specimens were tested under static tensile stress to extract some of the mechanical properties such as the ultimate strength and strain and the modulus of elasticity. The stress-strain data of all but one of the conditions (cooling rate and fiber orientation) were used to train the ANN and predict the stress-strain behavior for the remaining condition. The influence of ANN parameters such as type of ANN, number of hidden layers, number of neurons per hidden layer, and number of iteration of the network training on the prediction accuracy has been investigated. The best predictions were obtained by using a multilayer perceptron (MLPs) with two hidden layers and 50 neurons in each, both hidden layers were trained using RProp learning rule for 1000 epochs. For all of the cases investigated, the modulus of elasticity was predicted with a minimum accuracy of 97% while the ultimate strain was predicted, in most cases, with a minimum accuracy of 90%. These predictions indicate that ANN can be successfully used to predict the mechanical properties of unidirectional composites manufactured under different cooling rates.
    DSpace URI
    http://hdl.handle.net/11073/8760
    Collections
    • Masters Theses

    Browse

    All of DSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsCollege/DeptArchive ReferenceSeriesThis CollectionBy Issue DateAuthorsTitlesSubjectsCollege/DeptArchive ReferenceSeries

    My Account

    LoginRegister

    Statistics

    View Usage Statistics

    DSpace software copyright © 2002-2016  DuraSpace
    Submission Policies | Terms of Use | Takedown Policy | Privacy Policy | About Us | Contact Us | Send Feedback

    Return to AUS
    Theme by 
    Atmire NV