Project Title
Physics informed machine learning for smart monitoring and optimization of agricultural irrigation systems
Project Description
Agricultural irrigation systems play a crucial role in agri-food production and water resource management. Traditional methods face challenges in accurate spatiotemporal modeling, precise monitoring, and effective optimization due to soil heterogeneity, weather variability, and uneven water distribution. This project aims to develop physics-informed machine learning methods for smart monitoring of soil moisture and water usage optimization of agricultural irrigation systems. The candidate will leverage multi-modal data (time-series measurements and images) to develop predictive models and smart monitoring and control solutions for improving water-use efficiency. This project will be jointly supervised by Dr. Junyao Xie (Supervisor) and Dr. Fantahun Defersha (Co-supervisor).
Required Background
The candidate brings strong expertise in machine learning, including physics-informed neural networks, transfer learning, deep learning, and reinforcement learning. They have experience in modeling and optimizing agricultural irrigation systems or related fields, supported by solid programming skills in Python (using PyTorch and TensorFlow) and/or MATLAB. Their background includes numerical modeling, optimization, estimation, and control methods.
Additional Requirements
The candidate must be a Canadian citizen or permanent resident. However, international students with very strong academic background may be considered.
Funding
The position is fully funded according to the MASc stipend standards of the Department of Mechanical Engineering, College of Engineering, University of Guelph.
How to Apply
Interested applicants should email the following as a single PDF file to Dr. Junyao Xie (jxie10@uoguelph.ca):
· A brief cover letter describing your research interests and relevant experience
· An up-to-date CV (publications included if any)
· Academic transcripts (unofficial copies are acceptable at this stage)
Applications will be reviewed as they are received, and shortlisted applicants will be contacted for an interview. The position will remain open until filled.
Project Title
Machine learning-enabled smart monitoring of food manufacturing processes
Project Description
Food manufacturing involves complex thermal and mass transfer processes such as freezing, drying, and solidification, where multiphase transitions and moving boundaries play a critical role in determining product quality and safety. Ensuring efficient operation and quality control requires advanced condition monitoring systems. However, traditional model-based and data-driven approaches often face challenges in accurately capturing the spatiotemporal dynamics and moving boundaries of these processes. This project aims to develop machine learning-enabled methods for smart monitoring of food manufacturing processes. The candidate will integrate time-series sensor data and image information to design intelligent monitoring solutions that enhance process safety, reliability, and efficiency. This project will be jointly supervised by Dr. Junyao Xie (Supervisor) and Dr. Sheng Yang (Co-supervisor).
Required Background
The ideal candidate should have experience with machine learning algorithms such as Convolutional Neural Networks, Recurrent Neural Networks, and Long Short-Term Memory Networks. Experience in food manufacturing process modeling, monitoring, and optimization is highly desirable. Strong programming skills in Python (using frameworks such as PyTorch or TensorFlow) and/or MATLAB are required. The candidate must be a Canadian citizen or permanent resident.
Funding
The position is fully funded according to the MASc stipend standards of the Department of Mechanical Engineering, College of Engineering, University of Guelph.
How to Apply
Interested applicants should email the following as a single PDF file to Dr. Sheng Yang (syang19@uoguelph.ca) and Dr. Junyao Xie (jxie10@uoguelph.ca):
· A brief cover letter describing your research interests and relevant experience
· An up-to-date CV (publications included if any)
· Academic transcripts (unofficial copies are acceptable at this stage)
Applications will be reviewed as they are received, and shortlisted applicants will be contacted for an interview. The position will remain open until filled.
Project Title
Smart monitoring and diagnosis of distributed energy networks using generative AI
Project Description
Networked pipeline systems are one of the most cost-effective ways for long-distance energy transportation and transmission. With the global transition to a green energy era, existing pipeline infrastructures offer innovative and affordable pathways for transporting emerging energy carriers such as hydrogen, CO2, and ammonia. However, traditional monitoring and diagnosis methods face significant challenges due to complex network topologies, phase-change phenomena, and the lack of spatiotemporal monitoring data. This project aims to develop smart monitoring and diagnosis methods for state monitoring and abnormal detection (sensor/actuator faults, communication cyberattacks) of energy pipeline systems using generative AI. The candidate will develop image-based predictive models and AI-based smart monitoring framework to enhance the safety, reliability, and cybersecurity of next-generation energy transport networks. This project will be supervised by Dr. Junyao Xie.
Required Background
The candidate brings strong expertise in machine learning, including Convolutional Neural Networks, Variational Autoencoders, and Generative Adversarial Networks. Experience in modeling and monitoring of energy transportation systems or related fields is required, supported by solid programming skills in Python (using PyTorch, TensorFlow) and/or MATLAB. A strong background in numerical modeling, optimization, estimation, and control methods is also expected. The candidate must be a Canadian citizen or permanent resident.
Funding
The position is fully funded according to the MASc stipend standards of the Department of Mechanical Engineering, College of Engineering, University of Guelph.
How to Apply
Interested applicants should email the following as a single PDF file to Dr. Junyao Xie (jxie10@uoguelph.ca):
· A brief cover letter describing your research interests and relevant experience
· An up-to-date CV (publications included if any)
· Academic transcripts (unofficial copies are acceptable at this stage)
Applications will be reviewed as they are received, and shortlisted applicants will be contacted for an interview. The position will remain open until filled.