Enhancing Heat Transfer and Thermal Management Efficiency in Micro Heat Sinks through Machine Learning Algorithms - Doctoral Dissertation - October 2024

Aug 5, 2026

The growing power densities and rapid miniaturisation of modern electronics necessitate advanced thermal management solutions beyond conventional heat sinks, which struggle with limited performance and adaptability. This research develops a unified methodology combining machine learning, numerical simulations, and experimental investigations with manufacturing philosophies to set a new holistic benchmark. Bioinspired hybrid pin-fin geometries inspired by nature structures achieve 30–70% higher heat transfer potential, lower thermal resistance, and improved flow manipulation, while addressing manufacturability and thermal efficiency.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages280

Reading Time11 h 46 min

AccessFree to download

Abstract

Research aim and approach

Bioinspired micro heat sink designs

Reported performance improvements

Enhancing Heat Transfer and Thermal Management Efficiency in Micro Heat Sinks through Machine Learning Algorithms Mohammad Abdullah Hossain Harris Sustainable Engineering and Energy Technologies School of Physics, Engineering and Computer Science Supervisory Team: Professor Hongwei Wu (University of Hertfordshire) Dr. Anastasia Angelopoulou (University of Westminster) Dr. Wenbin Zhang (Nottingham Trent University) Examination Team: Professor Yuying Yan (University of Nottingham) Dr. Guogang Ren (University of Hertfordshire) Submitted to the University of Hertfordshire in partial fulfilment of the requirement of the degree of Doctor of Philosophy October 2024 Declaration of Originality I declare that the work in this dissertation was carried out under the requirements of the University’s Regulations and Code of Practice for Research Degree Programmes and has not been submitted for anyother academic award. Except where indicated by the specific reference in the text, the work is the candidate’s work. Any views expressed in the dissertation are those of the author. The following research output fully/directly contributed and was used in this PhD thesis: 1. Harris, M., Babar, H. and Wu, H., 2025. Assessing Thermohydraulic Performance in Novel Micro Pin-Fin Heat Sinks: A Synergistic Experimental, Agile Manufacturing, and Machine Learning Approach. International Journal of Heat and Mass Transfer, 239, p.126581 . 2. Harris, M., Angelopoulou, A., Wu, H. and Zhang, W., 2024, July. Comparative Analysis of Micro/Minichannel Flow Boiling Pattern Recognition and Classification using a Combined CNNClustering Algorithms Approach. In 2024 14th International Conference on Pattern Recognition Systems (ICPRS) (pp. 1-9). IEEE. 3. Harris, M., Wu, H., Angelopoulou, A., Zhang, W., Hu, Z. and Xie, Y., 2024. Heat transfer optimisation using novel biomorphic pin-fin heat sinks: An integrated approach via design for manufacturing, numerical simulation, and machine learning. Thermal Science and Engineering Progress, p.102606 . 4. Harris, M., Wu, H. and Sun, J., 2024, Investigating Heat Transfer and Flow Characteristics under Different Wall Heating Conditions in Novel Micro Pin-Fin Heat Sinks. in Proceedings ofthe 9th World Congress on Momentum, Heat and Mass Transfer (MHMT'24). 9th World Congress on Momentum, Heat and Mass Transfer (MHMT 2024), London, United Kingdom, 11/04/24 . 5. Harris, M. and Wu, H., 2023, Numerical Simulation of Heat Transfer Performance in Novel Biomorphic Pin-Fin Heat Sinks. in 8th International Conference on Experimental and Numerical Flow and Heat Transfer (ENFHT’ 23). 6. Harris, M., Wu, H., Zhang, W. and Angelopoulou, A., 2022. Overview of recent trends in microchannels for heat transfer and thermal management applications. Chemical Engineering and Processing-Process Intensification, 181, p.109155 . Other research output that partially/indirectly contributed to this PhD thesis: 7. Harris, M., Wu, H. and Toet, W., 2024. Reverse Engineering and Performance Enhancement of a NACA Duct for Cockpit Cooling in a GT4 Racing Car. In 4th SPECS Research Conference, University of Hertfordshire, Hatfield, UK (Accepted/In Press) . 8. Data Study Group Team., Harris, M., et al., 2022. In Data Study Group Final Report (Alan Turing Institute) . Casting the Future: understanding failure in turbine manufacture for Rolls Royce (Embargoed) . 9. Harris, M., 2021. An investigation on engine mass airflow sensor production via TQM, TPM, and Six Sigma practices. Operations Research Forum (Vol. 2, No. 4, p. 61) . Cham: Springer International Publishing. 10. Harris, M. Wu, H., Angelopoulou, A. and Zhang, W., Flow Boiling Regime Classification Using SIM-CLR Approach in Novel Hybrid Bioinspired Micro Pins. (Under Submission Process in the International Communications in Heat and Mass Transfer) 11. Harris, M. Wu, H., Sustainable Manufacturing of Micro Heat Sinks: A Life Cycle and Circular Economy Perspective.

FAQs

What core problem does the research address for microelectronics?

Conventional heat sinks cannot meet the demands of rapidly increasing power density and device miniaturisation, motivating more adaptive and efficient thermal management solutions.

What methodology does the thesis combine to evaluate micro heat sink performance?

The research integrates machine learning, numerical simulations, and experimental investigations while incorporating manufacturing philosophies into a single comprehensive workflow.

Which design strategy is used and what performance gains are reported?

Hybrid bioinspired pin-fin structures inspired by nature are modified strategically, yielding a 30–70% increase in heat transfer potential alongside reduced thermal resistance and improved flow control.

Aug 5, 2026

Upload by Seraphina