The increasing digitization of electric power systems has significantly expanded the attack surface of critical infrastructure, particularly within developing economies such as Nigeria, where grid modernization is still evolving. This study proposes an intelligent risk-aware security framework designed to detect, assess, and mitigate cyber attacks targeting critical power grid infrastructure in Nigeria’s electricity sector. The framework integrates threat intelligence, risk scoring, anomaly detection, and adaptive response mechanisms to enhance resilience against advanced persistent threats, including false data injection, GPS spoofing, SCADA manipulation, and IoT-based intrusions. Building on established cybersecurity models such as NIST cybersecurity framework principles and ENISA threat intelligence reports, the proposed system introduces a layered defense architecture tailored to smart grid environments. Machine learning-driven detection mechanisms, particularly deep learning models for anomaly recognition in phasor measurement units, are incorporated to improve detection accuracy (ALmutairy, Scekic & Wshah, 2022). The study synthesizes existing literature on cyber-physical power systems and identifies critical gaps in real-time risk prioritization and localized threat adaptation. The findings demonstrate that integrating risk-aware analytics with adaptive control significantly improves system resilience and reduces response latency. The research contributes a structured, scalable framework suitable for deployment in Nigeria’s power infrastructure and similar emerging grid environments.