Connecting urban energy systems through data
Xuan Liu defended her PhD thesis at the Department of Built Environment on June 10.
Urban energy transition involves many urban domains and diverse forms of urban data. Data heterogeneity is persistent across sources, scales, and indicator definitions. Data silos remain common in both research and practice. These conditions limit transparent comparison across places and constrain collective planning across organizational boundaries. Xuan Liu鈥檚 dissertation addressed this challenge by examining how semantic integration can enable the analysis of cross-sector interactions between building and transport energy systems in the context of urban energy transition.
The research began with a systematic review of district-level energy transition studies. The field has grown rapidly, but remains organized into separate technological, socio-political, governance, and market domains. This fragmentation highlights the need for an integrative analytical foundation.
Building on this insight, Liu鈥檚 dissertation proposes a neighborhood-level semantic structure for organizing urban energy data. It shows that data organization is not simply a technical step, but part of how urban energy problems are defined and understood.
Applications in solar energy and mobility
The proposed approach was applied in two cross-sector contexts. The first focused on neighborhood photovoltaic planning through the Neighborhood Photovoltaic Generation Ontology, linking PV system data with neighborhood characteristics, 3D building context, and time-based inputs. This made it possible to estimate hourly photovoltaic generation and compare different roof coverage scenarios, revealing both temporal variation and differences in solar potential.
The second application addressed the integration of building and transport energy systems through Electric Vehicle Charging Activity. By linking charging information, travel survey data, neighborhood characteristics, and photovoltaic capacity, the analysis revealed a temporal mismatch between electric vehicle charging demand and solar energy generation, as well as spatial differences in demand intensity and local supply鈥揹emand relations.
Enabling cross-sector urban energy analysis
These applications demonstrate that semantic integration can support measurable and planning-relevant analysis of interactions between buildings, transport systems, and local renewable energy sources. By connecting previously fragmented datasets, the approach enables a more comprehensive understanding of urban energy systems.
Key contributions of the research
The study concluded that digital integration through linked data provides a reusable foundation for cross-sector urban energy analysis. It makes three main contributions: identifying sectoral separation and heterogeneous data organization as key challenges; developing a neighborhood-scale semantic integration framework; and demonstrating its value through applications in photovoltaic generation and electric vehicle charging.
Overall, Liu鈥檚 dissertation shows that semantic integration can move beyond domain-specific data management and support interaction analysis across urban sectors.
Title of PhD thesis: Supervisors: Bige Tun莽er and Dujuan Yang.