The increasing complexity of modern urban energy systems necessitates robust and scalable data modeling frameworks to ensure efficient energy management across multiple levels. This paper presents a Hierarchical Energy Management System (HARM), a structured and flexible approach for energy data modeling that integrates buildings, neighborhoods, districts, and cities into a unified framework. By leveraging bottom-up data flow, top-down control mechanisms, and peer-to-peer (P2P) energy interactions, HARM proposes a new framework for seamless communication, real-time decision-making, and optimized resource allocation. The framework employs standardized data models, artificial intelligence (AI)-driven analytics, and Internet of Things (IoT)-enabled monitoring to enhance scalability, interoperability, and energy efficiency. A comparative analysis highlights the advantages of HARM over traditional top-down and bottom-up approaches, demonstrating its ability to balance autonomous decision-making with centralized coordination. The proposed model supports the integration of renewable energy sources, smart grids, and decentralized energy resources, making it a practical and adaptable solution for sustainable urban energy management.