Advanced Design Tools for Multi-Physics Metamaterials Using AI

Interdisciplinary Areas: Data and Engineering Applications, Future Manufacturing

Project Description

This project will explore the development of forward and inverse design tools for elastic metamaterials and metastructures leveraging the latest machine learning technologies. Metamaterials have opened new and exciting opportunities to achieve wave field manipulation (either electromagnetic, acoustic, or elastic) unattainable with traditional materials. However, the associated design space is virtually unbounded and the search for specific material functionalities has often proved to be a formidable and still largely unsolved task. This challenge becomes even more insurmountable when approaching the design of multi-physics metamaterials; examples include opto-acoustically coupled materials that have shown remarkable potential to enable recent concepts of quantum and topological metamaterials. With the advent of machine learning tools, it is possible to envision new design strategies that are specifically conceived to scour these large design spaces and leverage large amounts of data.

The postdoctoral researcher will perform theoretical, numerical, and experimental work at the interface of continuum mechanics, wave propagation, and artificial intelligence. The selected candidate will utilize the state-of-the-art facilities at Purdue’s Smart Informatix Laboratory, Structural Health Monitoring and Dynamics Laboratory, and Ray W. Herrick Laboratories in order to perform computations, material fabrication, and experiments.

Start Date

8/15/2025

Postdoc Qualifications

PhD in Mechanical Engineering, Civil Engineering, Aerospace Engineering, or closely related fields. Experience in computational and experimental continuum mechanics, Artificial Intelligence, Machine Learning, dynamics and vibrations is preferred.

Co-advisors

Mohammad R. Jahanshahi
Associate Professor of Civil Engineering
Associate Professor of Electrical and Computer Engineering (courtesy)
jahansha@purdue.edu
Smart Informatix Laboratory
http://web.ics.purdue.edu/~jahansha/people.html

Fabio Semperlotti
Professor of Mechanical Engineering
Professor of Aeronautics and Astronautics Engineering (courtesy)
fsemperl@purdue.edu
Structural Health Monitoring and Dynamics Laboratory
https://engineering.purdue.edu/~fsemperl/ 

Bibliography

1. Rih-Teng Wu, Mehdi Jokar, Mohammad R. Jahanshahi and Fabio Semperlotti, (2022), "A physics-constrained deep learning based approach for acoustic inverse scattering problems," Mechanical Systems and Signal Processing, 2022, Vol. 164, 108190, DOI: 10.1016/j.ymssp.2021.108190.
2. Rih-Teng Wu, Ting-Wei Liu, Mohammad R. Jahanshahi, Fabio Semperlotti, (2021), "Design of one-dimensional acoustic metamaterials using machine learning and cell concatenation," Structural and Multidisciplinary Optimization, 2021, Vol: 63, No. 5, 2399–2423, DOI: 10.1007/s00158-020-02819-6..
3. M. Jokar and F. Semperlotti, “Two-dimensional finite element network analysis: Formulation and static analysis of structural assemblies,” Comput. Struct., vol. 266, p. 106784, Jul. 2022.
4. S. Nair, T. F. Walsh, G. Pickrell, and F. Semperlotti, “GRIDS-Net: Inverse shape design and identification of scatterers via geometric regularization and physics-embedded deep learning,” Comput. Methods Appl. Mech. Eng., vol. 414, p. 116167, Sep. 2023.
5. Rih-Teng Wu and Mohammad R. Jahanshahi, (2019), "Deep convolutional neural networks for structural dynamic response estimation and system identification," Journal of Engineering Mechanics, American Society of Civil Engineers (ASCE), in press, Vol. 145, No. 1, January 2019, DOI: 10.1061/(ASCE)EM.1943-7889.0001556.