Abstract
Smart grid is evolving in different phases and artificial intelligence (AI) models are also the part of it. In the recent developments of Smart grid system there are research areas which are left unaddressed specifically in privacy techniques, scalability, more efficient data processing techniques. This Thesis highlights the research gaps on smart grid development across various integration phases. In the initial phase of Smart grid development, challenges included scalability issues in Machine-to-Machine (M2M) communication, communication delays affecting voltage stabilization and high computational costs in implementing Genetic Algorithm (GA)-based grid state estimation are covered in this review report. Review is further extended to Smart grid’s expansion phase which focused on 5G and AI integration, revealed difficulties in network reliability, latency, and electromagnetic interference-resistant protocols. In addition, the problems in implementing Convolutional Neural Networks (CNN) and Augmented Reality (AR) for grid operations are considered. Advancements in software-driven grid interoperability emphasized the need for edge controller integration and robust cyber-physical security measures to address privacy concerns, particularly with federated learning frameworks. This Thesis highlights the recent developments integrating blockchain, AI, and Multi-Agent Deep Reinforcement Learning (MADRL) uncovered complexities in optimizing real-time grid operations and ensuring secure, scalable communication. Emerging innovations such as Unmanned Aerial Vehicles (UAVs), 6G, and meta-learning models show promising results. There is a need of further exploration to address gaps in scalability, privacy-preserving techniques, and efficient data processing. These insights underscore the need for advanced, interoperable, and secure solutions to support the modernization and efficient operation of smart grids.
Publication Date
2025
Document Type
Thesis
Student Type
Graduate
Degree Name
Professional Studies (MS)
Department, Program, or Center
Graduate Programs & Research
Advisor
Abdulla Ismail
Advisor/Committee Member
Haris M. Khalid
Recommended Citation
Almheiri, Majid Jamal Majid Salim, "Machine Learning and Artificial Intelligence Models and Techniques in Modern Power Grid Communication Architecture" (2025). Thesis. Rochester Institute of Technology. Accessed from
https://repository.rit.edu/theses/12667
Campus
RIT Dubai
