About Me

Greetings! My name is Shutong Qi.

I am a Ph.D. candidate in Electrical and Computer Engineering at the University of Toronto, working at the intersection of scientific machine learning (SciML) and electromagnetic modeling. My advisor is Professor Costas D. Sarris.

From August to December 2024, I worked as an R&D intern with the Ansys Lumerical team, focusing on fast modeling and design automation of metalenses.

Before joining the University of Toronto, I received my bachelor’s degree in Electronics and Information Engineering from Beihang University in July 2020, where I was honored as an Excellent Graduate.

🔬 Research Overview

My research focuses on developing efficient, generalizable models for electromagnetic and multiphysics simulations using physics-informed machine learning (PIML), supporting applications in computational electromagnetics, high-speed circuit modeling, and automated simulation workflows.

🚀 Key Contributions

• Fast, Generalizable, Neural Operator-Based EM Solvers
  Developed a physics-informed Deep Operator Network (PI-DON) for 3D EM simulations, achieving up to 100× speedup over traditional FDTD methods. This enables efficient modeling for applications ranging from microwave circuits to metamaterials.

• Stable Time-Domain Modeling
  Proposed a hybrid PINN framework combining finite-difference time-stepping with neural-network-based spatial derivatives, allowing unconditionally stable wave simulations.

• Layout-to-Response Prediction
  Created a CNN-LSTM model to predict S-parameters directly from physical layout and substrate inputs. This supports rapid simulation and can assist in layout-aware design workflows, including signal integrity and optimization tasks.

• Multiphysics Simulation
  Built unsupervised PINN-U-Net architectures to simulate coupled electromagnetic and thermal processes, enabling data-efficient multiphysics modeling for reliability-aware system design.

My work is grounded in practical expertise across the full simulation workflow, from geometry and layout design to meshing, solver setup, post-processing, and visualization. I am proficient in tools such as Ansys HFSS, Lumerical, COMSOL, and CST Studio, and well-versed in scripting and automation for parametric sweeps, optimization, and results extraction.

💡 Recent News

• [12/2024] Our paper Physics-Informed Deep Operator Network for 3-D Time-Domain Electromagnetic Modeling has been accepted by the IEEE Transactions on Microwave Theory and Techniques. I would like to extend my sincere gratitude to my supervisor, Professor Sarris, for his support and guidance throughout the process.

• [12/2024] I recently completed my internship at Ansys Lumerical, where I worked on fast algorithms and adaptive Kriging modeling for metasurface design. I’m grateful for the valuable experience and would like to thank Dr. Jens Niegemann and all my colleagues for their support and guidance.

• [8/2024] I started my internship at Ansys Lumerical.

• [6/2024] I was honored to receive the MTT-S Graduate Fellowship and had the opportunity to present a coupled electromagnetic-thermal solver based on PINNs at IMS 2024.

• [1/2024] Our paper Hybrid Physics-Informed Neural Network for the Wave Equation with Unconditionally Stable Time-Stepping has been accepted by the Antennas and Wireless Propagation Letters.

• [1/2023] Our paper Electromagnetic-Thermal Analysis With FDTD and Physics-Informed Neural Networks has been accepted by the IEEE Journal on Multiscale and Multiphysics Computational Techniques.

• [10/2022] Our paper Deep Neural Networks for Rapid Simulation of Planar Microwave Circuits Based on their Layouts has been accepted by the IEEE Transaction on Microwave Theory and Techniques. Many thanks to my supervisor Professor Sarris.

• [09/2020] Start my PhD journey at University of Toronto.

• [07/2020] Graduate from Beihang University with an Excellent Graduate honor.