Artificial Intelligence for Grounding Systems in Smart Cities: A Systematic Review of Direct and Transferable Evidence for Design, Monitoring, Predictive Maintenance and Infrastructure Resilience

Grounding systems are essential components of electrical infrastructures, ensuring personnel safety, equipment protection, and operational reliability. In the context of smart cities, these systems have evolved from passive safety elements to critical assets within interconnected cyber-physical infrastructures. Simultaneously, advances in Artificial Intelligence (AI), the Internet of Things (IoT), digital twins, and predictive analytics have created new opportunities for improving the design, monitoring, maintenance, and resilience of electrical systems. This study presents a systematic review of AI applications in grounding systems and smart city electrical infrastructures following the PRISMA 2020 methodology. The literature search was conducted exclusively in the Scopus database, and 179 studies were included in the final synthesis. The review analyzes the scientific literature published between 2015–2026 and examines four major thematic classification dimensions: (i) the role of grounding systems in smart city infrastructures, (ii) AI-based design and optimization methods, (iii) intelligent monitoring, fault detection, predictive maintenance, and digital twins, and (iv) sustainability, resilience, and future implementation challenges. The thematic classification provides the organizational structure for the evidence synthesis, while a separate set of seven assessment axes is used to comparatively evaluate the technological capabilities and integration potential of the main AI technology families. The findings were synthesized by distinguishing between evidence directly addressing grounding systems and evidence from adjacent smart-grid and electrical-infrastructure domains. Direct grounding-system studies indicate that Machine Learning (ML), Deep Learning (DL), optimization algorithms, and intelligent sensing can support grounding design, condition assessment, fault diagnosis, and maintenance-related tasks, although the available evidence remains limited and heterogeneous. In adjacent electrical-infrastructure domains, AI-based approaches report high fault-detection performance, enable real-time condition monitoring, and support predictive maintenance strategies associated with reductions in downtime and maintenance costs. These findings provide transferable methodological insights for intelligent grounding-system management but should not be interpreted as direct evidence of performance improvements in grounding systems. The review also identifies critical research gaps, including the lack of standardized frameworks, limited large-scale deployments, interoperability challenges, cybersecurity concerns, and the need for explainable and trustworthy AI solutions. The study concludes that AI has the potential to become a foundational technology for the development of intelligent, adaptive, and resilient grounding systems that support the sustainability, reliability, and resilience objectives of future smart cities.