Hybrid Machine Learning Models for Enhancing Cybersecurity in Smart Grid Infrastructures

The increasing reliance of smart grid infrastructures on digital communication networks has made them highly vulnerable to cyber threats, particularly Distributed Denial-of-Service (DDoS) attacks. Traditional security mechanisms often struggle to detect and mitigate these sophisticated, evolving threats. This study proposes a hybrid machine learning model that enhances cybersecurity in smart grids by improving the accuracy and efficiency of DDoS attack detection and mitigation. The proposed model integrates supervised and unsupervised learning techniques, leveraging deep learning-based anomaly detection and ensemble classification algorithms to differentiate between normal and malicious network traffic in real-time. A comparative analysis of multiple machine learning classifiers, including Random Forest, Support Vector Machine (SVM), and Neural Networks, is conducted to assess performance in terms of detection accuracy, false positive rates, and computational efficiency. The model is evaluated using real-world and simulated datasets, demonstrating its ability to detect various types of DDoS attacks with high precision and minimal false alarms. By incorporating adaptive learning techniques, the model dynamically evolves to counter emerging cyber threats, ensuring robust security for smart grid communication networks. The results highlight the potential of hybrid machine learning approaches in reinforcing the resilience of next-generation smart grid infrastructures against cyber-attacks, thereby enhancing system reliability and stability.