Event Details
EV Charging Station Attack Detection Using Machine Learning
Presenter: Kamran Janwiri
Supervisor:
Date: Wed, December 11, 2024
Time: 09:30:00 - 00:00:00
Place: REMOTE Via Zoom
ABSTRACT
Abstract: As the adoption of Electric Vehicles (EVs) accelerates globally, with governments like Canada targeting 100% electric vehicle sales by 2035, the need for secure and reliable EV charging infrastructure becomes critical. EV charging stations (EVSEs) are increasingly targeted by cyberattacks such as reconnaissance, SYN floods, UDP floods, and backdoor intrusions, which can disrupt operations and compromise sensitive data.
This study explores the use of Machine Learning (ML) to enhance the security of EVSE systems. Using the CICEVSE2024 dataset and employs techniques such as Synthetic Minority Oversampling Technique (SMOTE) for data balancing and Principal Component Analysis (PCA) for feature selection. Multiple ML models, including Random Forest, k-Nearest Neighbors, and Gradient Boosting Machines, are evaluated to identify optimal solutions for cyberattack detection.
The findings demonstrate that Machine Learning can significantly improve EVSE security, ensuring robust, real-time threat detection while balancing performance and scalability. This work highlights the potential for ML to secure critical EV infrastructure, fostering confidence in the transition to electric mobility.
Join Zoom Meeting
Meeting ID: 741 4211 7065
Passcode: 599958