What it’s about
The retail sector is a substantial, largely untapped source of demand-side flexibility for smart grids: refrigeration, HVAC, lighting, and other store loads operate at scale and are reasonably predictable. Yet integrating retail buildings into smart-grid services is hampered by the heterogeneity of building automation systems, the lack of portable abstractions across vendors, and the cost of bespoke integration work for each new site or service.
This thesis investigates the software-architecture barriers behind that fragmentation and proposes a layered approach to address them:
- Empirical foundation – a bottom-up survey of four Danish retail store types documenting actual loads, control surfaces, and DR mobilisation potential.
- An Activity-Tracking Service that lets DR applications reason about why a load is on, not just that it is on.
- A Service Abstraction Layer (SAL) that decouples portable applications from vendor-specific control systems and introduces redundancy in service responsibility.
- Auto-configuring building services that reduce per-site integration cost.
- An ontology-based approach to discovering and reasoning about heterogeneous information available in modern Building Operating Systems.
Each contribution is grounded in a peer-reviewed paper across pilot retail sites and lab settings. Together they form a coherent stack: from understanding the empirical reality of retail loads, to giving applications the abstractions they need to be portable, to making BOS deployments cheap enough to deliver realistic ROI.
Supervision
- Principal supervisor: Mikkel Baun Kjærgaard
- Affiliation: Center for Energy Informatics, Faculty of Engineering, SDU
- Defended: August 2019
Citation
Hviid, J. (2019). A Software Approach to Mitigating Barriers for Smart Grid Integration in the Retail Sector. [PhD Thesis, SDU]. Syddansk Universitet. Det Tekniske Fakultet. https://doi.org/10.21996/4mb1-ec75
ISBN-13: 979-8870803661 (independently published print edition).