Search for a command to run...
This paper presents a conceptual and architecture-centric design study for Home Energy Management Systems (HEMS), introducing a cloud-based data and software engineering approach that emphasizes the organization, indexing, processing, and analysis of IoT-generated energy data. The proposed architecture supports scalable and reliable ingestion, storage, and retrieval of heterogeneous smart home data, laying the groundwork for high-performance analytics and real-time operations. Among the contributions, a conceptual framework is presented that compiles and classifies HEMS functionalities derived from various demand-side management programs and mechanisms, organized according to their relevance to key stakeholders, including end-users, service aggregators, and utility operators. This framework aims to leverage the potential benefits of these functionalities within multi-level energy communities, while also guiding the design of the proposed architecture and aligning its components with the operational, regulatory, and informational needs of each actor, thereby fostering dynamic interactions and user-centered service delivery. The architecture is structured into three interconnected environments-Ingestion, Operational, and Analytical-each responsible for enabling specific capabilities, from real-time monitoring and control to large-scale data analysis and decision support. By explicitly linking stakeholder needs with software components and data flows, the proposed system ensures adaptability, scalability, and meaningful participation in energy management. A conceptual evaluation demonstrates how the architecture supports representative HEMS use cases and stakeholder roles, offering a structured foundation for addressing emerging challenges in demand-side energy coordination and cloud-based HEMS architectures. Finally, the work includes the practical validation of the Ingestion Environment, providing experimental results that confirm the system’s scalability, performance, and reliability under realistic IoT workloads, thereby bridging the conceptual design with empirical evidence from an implemented component of the proposed architecture.