The evolution of microgrids has facilitated the integration of renewable energy resources into localized power systems, reducing reliance on non-renewable sources and enhancing energy sustainability. Among the various microgrid configurations, DC microgrids offer reduced power losses and increased operational efficiency, reliability, and flexibility. This paper proposes an intelligent machine learning-based system for DC microgrids, focusing on predicting load demands and automating resource redistribution to enhance energy management. The system also incorporates root cause analysis of anomalies, enabling proactive interventions to prevent system disruptions. By leveraging IoT and cloud-based technologies, real-time data from the microgrid is analyzed to optimize performance and secure reliability. A mobile application provides intuitive dashboards, anomaly notifications, and user-defined thresholds for operational control. Safety mechanisms such as relays and circuit breakers ensure prompt fault resolution, while predictive analytics support efficient energy distribution. This framework demonstrates the potential to revolutionize DC microgrid management through predictive and corrective intelligence.