Self-healing multi-agent techniques in electric power distribution systems: A review

Abstract

In the context of increasing complexity in power distribution networks, the development of self-healing systems has become essential. Interest and research in this area have grown significantly in recent years. The integration of self-healing capabilities has transformed the structure and operation of both conventional and smart distribution networks, particularly affecting protection and protection coordination systems. This paper presents a comprehensive review of the objectives, functions, applications, challenges, and recent advancements in self-healing technologies aimed at improving the control and operation of distribution systems. It also explores the technological and structural components that support self-healing management in smart grids. This paper focuses in particular on the role of self-healing in protection, protection coordination, fault location and isolation, system restoration, and power flow and voltage control. The study evaluates self-healing frameworks based on multi-agent systems for both control and protection in distribution networks. In these frameworks, primary protection relies on peer-to-peer communication among zone agents to detect and isolate faults efficiently. The main functions of these frameworks also include system monitoring, fault diagnosis and location, and effective grid restoration following fault isolation. Additionally, the paper reviews various self-healing control methodologies and strategies designed to enhance the performance of smart distribution systems, with a special focus on protection and coordination. The role of information and communication technologies, software tools, and measurement systems in achieving effective self-healing is also discussed. Furthermore, recent optimization techniques addressing self-healing challenges in distribution networks are examined. Finally, industrial challenges, such as communication delays and cyberattack threats, are discussed, followed by an analysis of existing research gaps and prospective directions for future work. A total of 147 relevant articles, including the most recent publications, have been analyzed. This review aims to serve as a rich and valuable source for those interested in developing the protection, control, operation, and reliability of power distribution systems through self-healing and multi-agent approaches.