Edge-AI Enabled Microgrids for Disaster-Resilient Smart Cities in Tropical Regions

Abstract

This study aims to develop and evaluate an Edge Artificial Intelligence (Edge-AI) enabled microgrid framework to enhance disaster resilience in smart cities located in tropical regions. A qualitative research approach was employed using a multiple case study design, as it enables an in-depth exploration of complex interactions between technological systems, environmental variability, and socio-institutional dynamics. The study was conducted in selected urban areas in Indonesia, representing typical tropical smart city environments exposed to climate-related disruptions. A total of twelve key informants were purposively selected based on their expertise in energy systems, urban planning, and disaster management, ensuring the richness and relevance of the data. Data were collected through semi-structured interviews, document analysis, and field observations, and analyzed using thematic analysis. The findings indicate that the integration of Edge-AI significantly improves real-time decision-making, adaptive energy management, and system autonomy, thereby strengthening microgrid resilience during disaster events. The results also highlight the importance of socio-technical alignment in ensuring successful implementation. This study recommends the adoption of decentralized intelligence architectures, enhanced stakeholder collaboration, and context-aware system design to optimize the performance and scalability of resilient energy infrastructures in tropical smart cities.