Artificial Intelligence Techniques for IoT-Driven Control and Monitoring of Substations and Smart Grids: Integration with Renewable Energy and Electric Vehicles using Holographic Convolutional Neural Network: Trends and Challenges

Abstract

The rapid evolution of smart grids requires advanced computational intelligence to ensure efficient, reliable, and secure power system operation. The convergence of Artificial Intelligence (AI) and Internet of Things (IoT) technologies enables intelligent substations with real-time monitoring, predictive analytics, and adaptive control capabilities. However, increasing renewable energy penetration and electric vehicle (EV) adoption introduce challenges involving load variability, grid stability, and cybersecurity. AI architectures, including Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, hybrid CNN-LSTM models, and emerging Holographic Convolutional Neural Networks (HCNNs), support accurate load forecasting, fault detection, demand management, and energy optimization. IoT sensors combined with cloud-edge computing facilitate continuous collection and processing of heterogeneous grid data. HCNNs further enhance high-dimensional representation, feature extraction, and pattern recognition across complex multi-source data streams. AI-based optimization also supports renewable integration and vehicle-to-grid operations, enabling EVs to participate actively in energy management. Furthermore, ensemble deep learning architectures strengthen cybersecurity in IoT-enabled grids. Despite these advances, challenges concerning scalability, data heterogeneity, privacy, security, and computational complexity remain. This review evaluates current AI-IoT methodologies, emerging trends, major challenges, and future directions for resilient, intelligent smart-grid systems.