Date of Award
1-1-2019
Document Type
Dissertation
Degree Name
Ph.D. in Engineering Science
First Advisor
Yacoub Najjar
Second Advisor
Hunain Alkhateb
School
University of Mississippi
Relational Format
dissertation/thesis
Abstract
In the first problem Polyetherimide graphene nanoplatelets papers (PEIGNP) were tested with different graphene loadings varying from 0-97 weight percent (WT%). The resulting stress-strain curves were utilized to develop two ANN models. Stress-controlled and strain-controlled models. Both models shoan excellent correlation to the experimental. Several Mechanical properties were calculated from the predicted stress-strain curves namely; toughness maximum strength maximum strain and maximum tangent modulus. Both models captured the same overall behavior of the PEIGNP composite. However the strain-controlled model was found to predict lower stress than the stress-controlled model. Finally a Graphical User Interface (GUI) was developed to aid in future use of the developed material. In the second problem a comprehensive investigation is performed to study the behavior of earthen embankments during an overtopping event. Due to experimental limitations numerical simulations are performed utilizing multi-phase Smoothed Particle Hydrodynamics (SPH) to study the post-failure behavior of the simulated embankments. This technique is validated by modeling different experiments focusing on various aspects of soil behavior such as; failure mechanism and seepage flow. Two hundred forty simulations are performed for different soil properties and embankment geometries. Embankment geometry consists of the side slope and height. The embankment slope range considered between 1.2:1 – 3:1 (H:V). And the height range is between 3-15 m. While the soil is divided into two sections; embankment and foundation soil. Four different soil types were considered for the embankment soil and five for the foundation soil. Many failure parameters were studied including; failure mode peak discharge Breach percent and initiation time and foundation erosion. Eight ANN models were developed to predict these failure parameters. The developed models shoan excellent correlation to the numerical simulations. Finally an EXCEL based GUI was designed to simplify the use of the developed models. Performance-based approach for material design risk assessment and emergency planning is utilized in this research. Artificial Neural Network (ANN) technique is employed to analyze and optimize two engineering problems; characterizing the stress-strain behavior of graphene nanocomposites and predicting earthen embankment failure due to overtopping. In the first application the optimization is based on experimental data and in the second application it is based on numerical data.
Recommended Citation
Qatu, Khalil, "Optimizing the performance of complex engineering systems aided by artificial neural networks" (2019). Electronic Theses and Dissertations. 1962.
https://egrove.olemiss.edu/etd/1962