AI-enhanced intrusion detection in smart renewable energy grids: A novel industry 4.0 cyber threat management approach
Umar Islam, Hanif Ullah, Naveed Khan, Kashif Saleem, Iftikhar Ahmad
Abstract
The rapid adoption of Industry 4.0 technologies in renewable energy grids has significantly improved efficiency and scalability. However, this integration has also amplified cybersecurity risks, making conventional Intrusion Detection Systems (IDS) insufficient against evolving cyber threats. This study proposes a novel AI-enhanced Intrusion Detection System (IDS) tailored for smart renewable energy grids, leveraging a multi-stage detection framework that integrates both supervised and unsupervised learning techniques. The proposed IDS combines Random Forest for signature-based detection and Autoencoders for anomaly-based threat identification, enabling real-time detection of both known and zero-day cyber threats. A comprehensive evaluation using real-world cyberattack datasets demonstrates that the system achieves a detection accuracy of 97.8 %, significantly reducing false positives compared to traditional IDS solutions. This work not only enhances the security and resilience of smart grids but also offers a scalable and adaptable cybersecurity framework for Industry 4.0 applications. The findings contribute to the advancement of AI-driven security mechanisms, ensuring the reliability of critical energy infrastructure in the face of sophisticated cyber threats.