Search

Browse

• Communities

My Space

• Login

Integration of innovative energy efficient technologies in buildings and neighborhoods

Lygerakis Filippos

Full record

---

Summary

This thesis develops and evaluates an innovative Knowledge Graph (KG)-based architecture integrated within a Digital Twin (DT) framework, designed to advance data and knowledge management capabilities at the building and neighborhood scale. Motivated by the limitations of traditional data handling, particularly in dealing with semantically fragmented and dynamic urban data, the proposed architecture leverages tailored ontologies that establish interconnected relationships and hierarchical structures. This enables semantic interoperability, contextualized reasoning, and integration of static and real-time data streams. The core innovation lies in uniting semantic KGs within a DT framework to support stakeholder-driven decision-making. Unlike conventional approaches that struggle with disconnected and incompatible data, the KG-DT architecture provides a structured, machine-interpretable model of building and neighborhood-level processes. It facilitates end-to-end data contextualization, advanced querying, and transparent analysis of retrofit scenarios. SPARQL-based querying is a central feature, allowing users to extract insights on material properties, energy performance, and scenario outcomes without manual filtering or simulation reruns. Two case studies demonstrate the approach's practical effectiveness. The first involved the integration of paraffin-based phase-change materials (PCMs) in gypsum and cement boards, combining experimental testing with EnergyPlus simulations. Results confirmed improved thermal performance and up to 22.3% energy savings for high-PCM gypsum boards, especially under hysteresis modeling. The second case study applied LCC and LCA analyses to evaluate 17 PV + battery scenarios at a university campus. The optimal configuration, combining bi-facial PV modules with lithium-ion battery storage under a self-consumption strategy, achieved the best balance between economic and environmental performance. Both cases required tailored ontologies: a lightweight material-focused model for Leaf House, and a modular, extensible structure based on Brick and SAREF for the TUC campus. The KG was populated using a combination of measured data, simulations, sensor metadata, and performance indicators. Overall, the KG-DT framework provides a scalable and semantically aligned decision support system, enabling transparent, data-driven planning for sustainable neighborhood renovations and future smart city applications.

Available Files

Services