Columbia EE Prof. Matthias Preindl and PhD Student Youssef A. Fahmy Win the Best Paper Award at MDPI Energies

Columbia University’s Matthias Preindl and PhD student Youssef A. Fahmy won the First Prize Best Paper Award at MDPI Energies for their innovative research on lithium-ion battery state-of-health estimation using convolutional neural networks together with McMaster University collaborators Ephrem Chemali, Phil Kollmeyer, and Ali Emadi.

By
Xintian Tina Wang
May 20, 2024

Columbia University's Electrical Engineering Department is proud to announce that Professor Matthias Preindl and PhD student Youssef A. Fahmy have won the prestigious Best Paper Award at MDPI Energies. Their groundbreaking research on state-of-health (SOH) estimation for lithium-ion batteries through convolutional neural networks (CNN) has garnered significant recognition for its innovative approach and practical implications. Selected from over 9000 papers published by Energies in 2022, the paper Professor Matthias Preindl and PhD student Youssef A. Fahmy worked on has come to the top and won "The First Prize." The paper is based on research conducted in collaboration with McMaster University.

Award-Winning Research: A Novel CNN-Based Framework for Battery SOH Estimation

The award-winning paper, titled "A Convolutional Neural Network Approach for Estimation of Li-Ion Battery State of Health from Charge Profiles,” addresses a critical challenge in the field of battery management systems. As lithium-ion batteries become increasingly ubiquitous in applications such as electric vehicles, smart grids, and various electronic devices, accurately estimating their state-of-health is essential for ensuring safety and reliability.

The research introduces a CNN-based framework for estimating the SOH of lithium-ion batteries using data collected during the charging process. The CNN model was trained using data from 28 cells aged at two different temperatures with randomized usage profiles. The network's architecture ranged from 1 to 6 layers with 32 to 256 neurons, and the training data was augmented with noise to improve accuracy.

Key findings include:

  • The proposed CNN achieved a mean average error (MAE) as low as 0.8% over the battery's life.
  • Even with partial charge data, the CNN maintained a reasonable MAE of 1.6%.
  • The model's performance was validated for various state-of-charge (SOC) ranges, demonstrating its robustness for practical applications.

The research stands out for its ability to estimate SOH accurately without relying on physical or electrochemical models, which are often complex and challenging to develop. This data-driven approach, leveraging deep learning, marks a significant advancement in battery management systems, promising enhanced performance and reliability for various battery-powered technologies.

Matthias Preindl: An Acclaimed Expert in Electrical Engineering

Dr. Matthias Preindl, a prominent figure in the field of electrical engineering, has a rich academic and professional background. He received his B.Sc. degree from the University of Padua with summa cum laude honors in 2008, followed by an M.Sc. degree from ETH Zurich in 2010, and a Ph.D. from the University of Padua in 2014. Since 2016, he has been an Associate Professor at Columbia University.

Before his tenure at Columbia, Dr. Preindl worked as an R&D Engineer at Leitwind AG, specializing in power electronics and drives. He also served as a Post-Doctoral Research Associate at McMaster University in Canada. A Senior Member of IEEE and Fellow of IET, Dr. Preindl has contributed extensively to the field, serving as Area Editor of IEEE Transactions on Vehicular Technology and Treasurer of the IEEE Transportation Electrification Council (TEC). His research interests include the design and control of motor drives, power electronics, and batteries for transportation electrification.

Youssef A. Fahmy: A Rising Star in Power Electronics Research

Youssef A. Fahmy is currently a PhD student at Columbia University, working in the Power Electronics and Motor Drives Laboratory under the supervision of Dr. Preindl. Fahmy holds dual B.Sc. degrees in Physics from Brandeis University and Electrical Engineering from Columbia University, both obtained in 2020, followed by an M.Sc. degree in Electrical Engineering from Columbia University in 2022. His research focuses on power electronics, electric vehicles, and alternative energy resources.

The recognition of Professor Matthias Preindl and Youssef A. Fahmy with the Best Paper Award at MDPI underscores the innovative and impactful nature of their research. Their work not only advances the field of electrical engineering but also holds great promise for the future of energy storage and management technologies. Columbia University congratulates them on this remarkable achievement and looks forward to their continued contributions to the field.

Read full paper here.