Transparent Management of Dynamically Evolving Data in Smart Local Energy Systems-By means of Knowledge Graphs

Abstract

Smart Energy Systems represent a shift from vertically oriented approaches within the limits of sub-sectors/infrastructures in the area of generation, transmission and distribution of electricity into an integrated approach that leads to coherent sustainable energy systems. As a result of this transition, new subsystems – such as buildings that both generate and consume energy, transport and mobility systems incorporating electric vehicles, and community
or individual energy generation from renewable sources and engagement – also become part of the energy system. This enhanced scope results in the emergence of a diversified way of describing data pertaining to the system elements. When the structure of data and system elements within such an energy system exhibits dynamic behaviour (e.g. by adding new elements,
new or changed data formats etc.), maintaining a consistent functionality of dependant services becomes a challenge. This is particularly relevant for functionality delivered through distributed ledger technologies, where any recorded data (including its structure and format) must remain unchanged in order to be processable. The elements of the distributed system that handle data collection and processing can utilise a semantic layer to help them decouple the consistent service delivery from the evolving, dynamic state of the underlying data sources. In this thesis, the author presents the design and development aspects of such a layer – a distributed middleware system called Prototype-B – that supports the consistent functionality of services in
distributed system applications despite the presence of ongoing evolutionary change and heterogeneously described data. Prototype-B does this by transparently maintaining a dynamic ontology that hierarchically arranges semantic structures abstractly describing elements of the system, oversees their evolution and validates datasets against them. When a change occurs, Prototype-B validates previously used and newly suggested datasets against the agreed semantic structures. Finally, when requested, it helps services and applications identify required data sources and obtain data from them.

Date of Award	3 Oct 2023
Original language	English
Awarding Institution	University of Bristol
Supervisor	Ruzanna Chitchyan (Supervisor)