Physics-Informed Machine Learning Methods for Inverse Design of Multi-Phase Materials with Targeted Mechanical Properties - Dissertation

Aug 3, 2026

Upload by Ezra

Advances in machine learning have strengthened engineering inverse design, enabling microstructures to be searched or generated for desired mechanical behavior. This dissertation develops physics-informed predictive and generative neural networks for inverse design of multi-phase materials with targeted linear and nonlinear mechanical properties in both 2D and 3D settings. For 2D porous materials, ResNet predicts elastic modulus and stress-strain curves from microstructure images, while VAE-based models generate images from prescribed curves with mechanics enforced through a new condition fusion layer and physically meaningful constraints. For 3D fiber-reinforced polymer composites, a physics-informed diffusion model reconstructs and generates feasible fiber distributions under non-collision constraints, producing tailored mechanical behaviors.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages136

Reading Time5 h 43 min

AccessFree to download

INTRODUCTION AND MOTIVATION

Material Inverse Design

Machine Learning and Inverse Design

Research Questions and Approach

Clemson University TigerPrints | All Dissertations | Dissertations | | --- | --- | | 8-2024
Physics-Informed Machine Learning Methods for Inverse Design of Multi-Phase Materials with Targeted Mechanical Properties
Yunpeng Wu
[yunpeng@g.clemson.edu](yunpeng@g.clemson.edu)
Follow this and additional works at: [https://open.clemson.edu/all_dissertations](https://open.clemson.edu/all_dissertations)
Part of the Artificial Intelligence and Robotics Commons, Data Science Commons, Polymer and Organic Materials Commons, and the Structural Materials Commons | | Recommended Citation Wu, Yunpeng, "Physics-Informed Machine Learning Methods for Inverse Design of Multi-Phase Materials with Targeted Mechanical Properties" (2024) . All Dissertations. 3657. [https://open.clemson.edu/all_dissertations/3657](https://open.clemson.edu/all_dissertations/3657) This Dissertation is brought to you for free and open access by the Dissertations at TigerPrints. It has been accepted for inclusion in All Dissertations by an authorized administrator of TigerPrints. For more information, please contact [kokeefe@clemson.edu](kokeefe@clemson.edu). PHYSICS-INFORMED MACHINE LEARNING METHODS FOR INVERSE DESIGN OF MULTI PHASE MATERIALS WITH TARGETED MECHANICAL PROPERTIES A Dissertation Presented to the Graduate School of Clemson University In Partial Fulfillment Of the Requirements for the Degree Doctor of Philosophy Mechanical Engineering By Yunpeng Wu August 2024 Accepted by: Dr. Gang Li, Committee Chair Dr. Feng Luo Dr. Huijuan Zhao Dr. Oliver Myers ABSTRACT Advances in machine learning algorithms and applications have significantly enhanced engineering inverse design capabilities. This work focuses on the machine learning-based inverse design of material microstructures with targeted linear and nonlinear mechanical properties. It involves developing and applying predictive and generative physics-informed neural networks for both 2D and 3D multiphase materials. The first investigation aims to develop a machine learning method for the inverse design of 2Dmultiphase materials, particularly porous materials. We first develop machine learning methods to understand the implicit relationship between a material's microstructure and its mechanical behavior. Specifically, we use ResNet-based models to predict the elastic modulus and stress-strain curves of linear and nonlinear porous materials from their microstructure images. To generate microstructures of porous materials with targeted mechanical behavior, we create variational autoencoder (VAE) based neural networks. These networks generate the microstructure of porous materials from a prescribed elastic modulus or stress-strain curve. In both property prediction and microstructure generation, the stress-strain curves are approximated using cubic polynomials and characterized by their coefficients. To explicitly enforce the mechanics of materials in the generative machine learning models, we devise and incorporate a new condition fusion layer into the traditional VAE architecture. Additionally, a pretrained regression model is introduced to constrain the decoder, ensuring the production of physically meaningful images. The results show that this machine learning approach is capable of ultra-fast prediction of material properties directly from microstructure images, as well as the inverse design of material microstructures to achieve desirable mechanical behaviors. The second investigation focuses on the inverse design of 3D multiphase materials, specifically considering fiber-reinforced polymer composites (FRPC) as the model system. This research aims to develop physics-informed neural networks for inverse design of such a material system. Compared to 2D porous materials, 3D FRPC involve complex 3D microstructure geometries and require physically feasible topologies, making the inverse design significantly more challenging. To address these challenges, we develop

FAQs

What problem does this dissertation address?

It addresses machine learning-based inverse design of material microstructures to achieve targeted linear and nonlinear mechanical properties.

How are 2D porous materials handled?

ResNet-based models predict elastic modulus and stress-strain behavior from microstructure images, and VAE-based networks generate microstructures from prescribed mechanical responses while enforcing mechanics with a condition fusion layer and constraints.

What approach is used for 3D fiber-reinforced polymer composites?

A physics-informed diffusion model reconstructs and generates 3D fiber distributions using a stochastic differential equation formulation, with non-collision constraints to ensure feasible topology and tailored mechanical behavior.

Aug 3, 2026

Upload by Ezra