Faculty/Department: Engineering/Mechanical
Project Title: Physics-Informed Machine Learning Control for Hydrogen Maximization in Heavy-duty Engines
Position Type: Doctoral Student
About the Position
The University of Alberta invites applications for a Doctoral (PhD) position in Mechanical Engineering within the Mechanical Engineering Energy Control Lab (MEECL). This role focuses on the integration of Physics-Informed Neural Networks (PINNs) and Kolmogorov-Arnold Networks (KANs) to optimize hydrogen combustion engines for heavy-duty transport. This PhD opportunity is for international students who wants to join study in Canada
Purpose
The person hired into this position will be a PhD student who will work with the Mechanical Engineering Energy Control Lab (MEECL) team to investigate the integration of new and promising ML networks for engineering applications. Specifically, Physics Informed Neural Networks (PINNS) allow for the integration of physics and ML to improve the prediction performance and extrapolation capabilities of the model used within Model Predictive Control (MPC), while Kolmogorov-Arnold-Networks (KANs) provide a new network structure that greatly outperforms traditional neural networks in terms of accuracy while using small networks. Both PINNs and KANs allow engineers to integrate physics back into ML and are expected to provide significant real-time controller benefits as the developed networks are simpler, thus requiring reduced computational resources and less training data to help facilitate future commercial products. Thus, this project will look at integrating these new ML methods for hydrogen internal combustion engine control. These controllers will be experimentally validated in the recently completed testing facility, which provides cutting-edge testing capabilities for the performance and emissions of heavy-duty transport engines and testing of the developed ML-based controllers.
The successful candidate will be a part of the low-carbon transportation research team within the Department of Mechanical Engineering. The successful applicant will be expected to complete research, analyze results, and present results in conferences, journal publications, and at internal meetings to industry partners.
Project Overview
The research aims to:
- Improve real-time controller performance using ML and physics-based models
- Enhance model predictive control (MPC) systems for hydrogen engines
- Develop low-carbon transportation solutions
- Conduct experimental validation in a state-of-the-art testing facility
Required Skills & Qualifications
š MSc in Mechanical Engineering or a related field
š Strong background in control system design & implementation
š Experience with Machine Learning (ML) integration
š Knowledge of PINNs and KANs (preferred)
š Proficiency in Python, C Code, Matlab/Simulink
š Hands-on experience with experimental controller integration
š Research experience with optimal control strategies and data analysis
Why Apply?
ā Be part of a cutting-edge low-carbon transportation research team
ā Work with leading experts in ML-based control systems
ā Conduct experimental research using world-class testing facilities
ā Publish in high-impact journals and conferences
Application Process
š© Send a single PDF file with the following documents to dgordon@ualberta.ca with the subject line:
“Prospective PhD StudentāPINN for H2DF Enginesā[Your Full Name]”
š Required Documents:
- Ā Cover Letter (maximum 2 pages) summarizing research experience & interests
- Academic CV (including publications & references)
- Transcripts of prior degrees
- For international applicants: English language test results (e.g. TOEFL, IELTS)
Location & Start Date
š University of Alberta, North Campus (Edmonton, AB)
š Start Date: September 1, 2025 (or when a suitable candidate is found)
š Program link: