Fuzzy Clustering and Autoencoder-Based Cybersecurity for EVs in a Blockchain-Enabled Smart Cloud Environment

The rise of electric vehicles (EVs) has revealed several cybersecurity weaknesses, especially in data transmission inside cloud-based systems. Current cybersecurity algorithms for anomaly detection frequently struggle to manage the intricacy and magnitude of electric vehicle network data. This study presents an innovative cybersecurity system that integrates fuzzy clustering for soft anomaly detection, autoencoders for deep learning-driven feature extraction, and blockchain for secure data validation. The suggested model effectively tackles critical issues in anomaly detection, attaining an accuracy rate of 97.2%, an F1 score of 0.96, and a recall of 0.95, therefore surpassing current methodologies substantially. The results underscore the model's potential to identify and address cyber risks with enhanced precision and recall, providing a scalable and resilient solution for EV networks. Future improvements may concentrate on minimising computation time and augmenting detection efficacy in real-time systems, establishing this method as a viable option for bolstering EV cybersecurity in blockchain-integrated smart settings.