Computer Vision and Machine Learning Enabled Prediction of Two-Phase Flow and Heat Transfer
Project Description
Thermal management systems that leverage liquid-to-vapor phase change are critical for applications ranging from large-scale AI computing to electric vehicle traction drives. However, the design and optimization of systems that can achieve these benefits in practice is hindered by the inability to model and predict two-phase flow and heat transfer with high certainty and generality. To address this gap, emerging physics-informed (PhI) machine learning (ML) and computer vision techniques have been shown to enable interpretation and forecasting of noisy and chaotic dynamic systems, as well as leveraging image interpolation/extrapolation to extract/quantify salient physical features. The Lillian Gilbreth Postdoctoral Fellow would join our METHODS (Machine learning Enabled Two-pHase flow metrologies, models, and Optimized DesignS) MURI (Multidisciplinary University Research Initiative) program, which brings together a multidisciplinary team of experts to enhance fundamental understanding and enable generalized prediction of phase change during two-phase flows. The Gilbreth Fellow will explore techniques that leverage PhI ML and computer vision techniques to enhance understanding two-phase flow boiling (spanning potential topics on 3D vision and scene reconstruction, generative AI, object segmentation and tracking, etc.).
Start Date
2026
Postdoc Qualifications
Background in electrical engineering or computer science with expertise in computer vision and image processing. Experience in physics-informed machine learning, modeling/metrology expertise related to phase change processes, interfacial transport phenomena, or flow boiling heat transfer is also desired.
Co-advisors
Justin A Weibel, jaweibel@purdue.edu Professor, School of Mechanical Engineering; Director, Cooling Technologies Research Center: https://engineering.purdue.edu/CTRC/research/index.php
Fengqing Maggie Zhu, zhu0@purdue.edu Associate Professor, Elmore Family School of Electrical and Computer Engineering; https://engineering.purdue.edu/~zhu0/
Bibliography
- Y. Shen, Q. Zhai, F. Zhu. PS4PRO: Pixel-to-pixel Supervision for Photorealistic Rendering and Optimization. Proceedings of the Conference on Computer Vision and Pattern Recognition Workshop, Jun 2025.
- J.H. West, A. Ceperley, P.V. Vydylua, and J.A. Weibel, Synchronous through-substrate high-speed visual and infrared observation of flow boiling in a rectangular channel, ASME Summer Heat Transfer Conference (SHTC), Colorado, USA, July 8-10, 2025.
- K.N.R. Sinha, P.V. Vydylua, M. Bongarala, J.A. Weibel, A deep learning approach for heat flux partitioning analysis of pool boiling using through-substrate infrared thermography, ASME Summer Heat Transfer Conference (SHTC), Colorado, USA, July 8-10, 2025.
- X. Ji, A. Shakouri, F. Zhu, Confidence-Aware Agglomeration Classification and Segmentation of 2D Microscopic Food Crystal Images, Proceedings of the IEEE International Conference on Image Processing, Sep 2025.C. Zhang, Y. Shen, F. Zhu, ICP-3DGS: SfM-Free 3D Gaussian Splatting for Large-Scale Unbounded Scenes. Proceedings of the IEEE International Conference on Image Processing, Sep 2025.