Machine learning for modeling ice growth dynamics and crystal morphology
Thomas Schutzius, Professor
Mechanical Engineering
Applications for Fall 2025 are closed for this project.
This project aims to apply supervised machine learning to understand and predict the ice growth process in various solutions. Beyond modeling the overall freezing rate, we seek to capture detailed aspects of ice crystal morphology — such as tip radius, dendrite arm spacing, and branching patterns — under different thermal and chemical conditions. By training models on experimental and/or simulation data, we aim to link the physical properties of solutions (e.g., density, melting point, polymer molecular weight, viscosity) with the resulting ice growth rate and morphological features. This research contributes to applications in cryopreservation, freeze desalination, and materials processing.
Qualifications: The undergraduate researcher will work on collecting, organizing, and preprocessing experimental or simulation data on ice growth rates and crystal morphology. They will help train and validate supervised machine learning models to find relationships between solution properties and ice growth behaviors. Tasks include data cleaning, feature engineering, testing various algorithms, and visualizing results. The student will gain valuable experience in applying machine learning to solve complex thermal science problems, understand phase-change processes, and develop skills in scientific data handling and model development.
Day-to-day supervisor for this project: Shuai Li, Post-Doc
Hours: 12 or more hours
Off-Campus Research Site: Required: Basic programming knowledge (Python preferred) Interest in thermal sciences and phase-change phenomena Strong analytical thinking and willingness to learn machine learning techniques Desirable but not essential: Familiarity with supervised machine learning methods Experience with Python libraries such as Pandas, Scikit-learn, or TensorFlow Background coursework in thermodynamics, heat transfer, or fluid mechanics Class level / major: Open to all undergraduate levels; students majoring in Mechanical Engineering, Chemical Engineering, Materials Science, or related disciplines are encouraged to apply.
Related website: https://