Expert and Intelligent Systems for Assessment and Mitigation of Cascading Failures in Smart Grids: Research Challenges and Survey

Network instability conditions in a smart grid may lead the network to cascading failure events (CFEs) which ultimately lead to blackouts. Examination of these CFEs at an early stage will help the network operators to mitigate the further propagation of these events in the power system network. There are many artificial intelligence-based topologies to identify, analyze, and prevent these types of events. Selecting an appropriate topology by looking at the power network architecture is one of the critical research issues that needs to be resolved. For this purpose, this review study provides a thorough examination and evaluation of intelligent assessment methodologies in smart grids to avoid CFEs, including an exploration of both their advantages and shortcomings. In contrast to existing review studies, this research focuses on a wide range of advanced topologies, i.e., quasi-steady state methods, dynamic methods, artificial intelligence (AI), probabilistic approach, digital twin method, blockchain techniques, metaverse, and the advance control methods in smart grids to avoid CFEs. Similarly, the mitigation strategy we highlighted includes several optimal power flow algorithms based on advanced machine learning that can be integrated into smart grid infrastructure to compensate against CFEs in smart grids. The objective is to pave a decision-making path for the scientific researchers who want to contribute to this research area. Through a comparative analysis of a wide range of these cascading failures assessment and mitigation topologies, the network operators easily identify the proactive approaches that can be utilized at the early stages to detect and mitigate cascading failure vulnerabilities, thus ensuring the resilience of the smart grid. Moreover, this research work indicates areas where further research is needed, and suggesting potential directions for future investigations.