AI-Assisted Composite Materials Design

About This Course

“AI-Assisted Composite Materials Design” is an immersive 3-week course that explores how Artificial Intelligence (AI) is revolutionizing traditional materials science workflows. In this course, participants will learn how to apply data-driven models, surrogate optimization, and deep learning algorithms to predict material properties, simulate behaviors, and discover new composite formulations with tailored mechanical, thermal, or electrical properties.

The course emphasizes real-world datasets, multi-scale modeling, and AI-powered tools like Bayesian optimization, Neural Networks, Graph Neural Networks (GNNs), and AutoML platforms applied to composite design and simulation.

Aim

The aim of this course is to equip participants with the knowledge and tools necessary to leverage AI and Machine Learning (ML) for the design, modeling, and optimization of composite materials. By doing so, participants will be empowered to accelerate innovation across aerospace, automotive, energy, and biomedical applications.

Course Objectives

• Bridge the gap between materials science and artificial intelligence

• Train participants to use AI for faster, cost-effective materials discovery

• Foster cross-disciplinary collaboration for smart, sustainable material development

• Introduce scalable digital tools for next-generation composite design

• Promote reproducibility, transparency, and innovation in AI-assisted materials research

Course Structure

📅 Module 1: Generative Models for Microstructure Design

Theme: Harnessing AI for Microstructure and Property Design

• Fundamentals of Microstructure Design

  • Understanding the impact of microstructure on material properties

  • Traditional vs. data-driven design approaches

• Overview of Generative Models

  • GANs (Generative Adversarial Networks), VAEs (Variational Autoencoders), and Diffusion models

  • Conditioning generative models on target properties

• Learning Inverse Design

  • Mapping structure to desired material properties

Hands-On Lab:

• Case studies in 2D/3D material generation using machine learning

• Exploring generative models for real-world applications

📅 Module 2: Bayesian Optimization for Stiffness/Weight Trade-Off

Theme: Optimization Techniques for Composite Materials

• Multi-Objective Design Problems

  • Exploring trade-offs like stiffness vs. weight in materials and components

• Bayesian Optimization

  • Gaussian processes, surrogate models, and acquisition functions

  • Pareto frontiers and uncertainty quantification

Hands-On Lab:

• Practical optimization using Bayesian techniques for composite design

📅 Module 3: Digital Twin Validation in Finite Element Analysis (FEA)

Theme: Advanced Simulation Techniques for Material Design

• Introduction to Digital Twins

  • How digital twins are used in predictive engineering

• Integrating Simulation Data with Real-World Observations

  • Setting up and validating FEA models for structural behavior

• AI-Assisted Model Calibration

  • Techniques for enhancing model fidelity and simulation performance feedback

Hands-On Lab:

• AI-assisted model updates to improve simulation accuracy

• Validating FEA models with experimental data

Who Should Enrol?

• Materials and mechanical engineers

• Polymer scientists and nanocomposite researchers

• AI/ML engineers in manufacturing or R&D

• Aerospace, automotive, and biomedical materials developers

• UG/PG/PhD students in materials science, physics, or applied AI