Graduation Semester and Year

Summer 2025

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Industrial Engineering

Department

Industrial and Manufacturing Systems Engineering

First Advisor

Bangde Liu

Abstract

Fiber-reinforced composites (FRCs) are widely used in aerospace, automotive, and other engineering applications due to their lightweight characteristics and superior mechanical properties. Traditional FRCs employ a fixed fiber orientation in each layer, and while different layup sequences can tailor performance, the mechanical properties remain spatially uniform. Tow-steered composites, which enable fibers to follow curvilinear paths, offer the potential for spatially varying stiffness and strength, improving structural performance.

This dissertation addresses key challenges in modeling and designing tow-steered composite structures, including the lack of commercial design tools, the high computational cost of design optimization, and the need to efficiently evaluate manufacturing defects. To address the above challenges, the following contributions are made:

  1. Development of a Design Tool for Advanced Tailorable Composites (DATC): This tool is developed as graphical user interface (GUI) plugins into commercial finite element (FE) software Abaqus and MSC.Patran/Nastran, integrating structural analysis, plate constitutive modeling, and design optimization into a unified framework for tow-steered composite structures.
  2. Development of transfer learning models and new adaptive sampling algorithms: These models and algorithms significantly reduce computational costs when generating training data for constructing efficient surrogate models. The surrogate models offer an ultra-efficient alternative to FE-based analysis and design of tow-steered composite structures.
  3. Development of two-step, physics-guided convolutional neural network (CNN) models for defect evaluation: These models can rapidly compute the entire stress-strain curves of composite laminates with fiber misalignment, considering complex fiber-kinking failure behaviors.

The DATC tool has been validated on a range of aerospace structures, demonstrating its capability to streamline the design process and reduce manual programming. A machine learning (ML) module embedded within DATC enables NN-based optimization with high accuracy and substantial computational savings compared to traditional FE-based methods. Furthermore, transfer learning models and adaptive sampling are applied to several tow-steered composite structures to demonstrate the cost reduction for developing efficient surrogate models. The CNN models have been developed to capture fiber misalignment and predict nonlinear stress-strain curves of composite laminates under compression, effectively capturing the influence of local defects. This dissertation has resulted in several journal and conference papers and a design tool package.

Keywords

Tow-steered composites, Machine learning, Neural networks, Design optimization, Multiscale modeling

Disciplines

Manufacturing | Systems Engineering and Multidisciplinary Design Optimization

Download

Available for download on Tuesday, August 11, 2026

Share

COinS