Renewable Energy Sources (RES) are increasingly penetrating the energy production side, and in combination with DR programs and improvement in energy storage options, could contribute to significantly reduce peak demands. However, the integration of the different systems and technologies involved in the distributed energy consumption and generation is a big challenge.
The different pilot dwellings of the RESPOND project have different devices deployed including sensors, appliances, meters or energy assets produced by different manufacturers. Over the years, device manufacturers have embraced different communication technologies that are mutually incompatible. As a matter of fact, there are cases where even devices using same communication technology (e.g. ZigBee) produced by different manufacturers are mutually incompatible, impeding the easy upgrade of the system with new devices or features. In addition, data coming from external systems such as weather forecasting systems or aggregators will also be leveraged in RESPOND to perform different activities such as energy data analytics to detect potential energy conservation opportunities. Having such a variety of data sources poses a big integration challenge, not only because every data source has its own data model, but also because the same concept may be defined in different ways. Furthermore, the interoperability is another hurdle to be solved.
Interoperability is defined as the ability of a system to work with or use the parts of equipment of another system. There exist three interoperability layers, as shown in the following image: technical interoperability, syntactic interoperability and semantic interoperability.
Semantic Technologies can be leveraged to remedy the aforementioned issue as they enable integrating data across several data sources. More specifically, ontologies are foreseen as the main drivers to address this challenge. An ontology can be defined as a formal, explicit specification of a shared conceptualisation. The conceptualisation specified by each ontology is usually devoted to representing a certain phenomenon, topic, or subject area, and designed with a certain purpose.
The RESPOND ontology has been created as a solution for the integration of heterogeneous data available in the RESPOND project. Its development has followed the good practices of modularity and reuse and it has been guided by the well-known NeOn Methodology. A study of related domain ontologies has been conducted and some of them have been selected to reuse based on a conceptual agreement with the RESPOND requirements, axiomatic richness relating their terms, simplicity of the structure to facilitate querying, popularity of the ontology to improve interoperability, and documentation accessibility to facilitate new users.
As a matter of fact, RESPOND ontology’s core is built by reusing and extending three well-known ontologies: BOT to represent the dwelling topology, and SAREF and SEAS Feature Of Interest ontologies to represent devices, features of interest and qualities monitored and controlled by sensors and smart appliances. Furthermore, parts of other ontologies and vocabularies have been reused and extended to represent units of measurements (QUDT Unit) and different qualities (M3-Lite taxonomy). The following figure shows the main classes and properties defined in the RESPOND ontology.
The RESPOND ontology is leveraged to semantically annotate data coming from different pilot sites. Semantic Annotation is the process of linking existing syntactic information with specific ontologies to provide both machine understandable and human readable descriptions. Ontologies provide semantics to existing data and furthermore link different information together via predefined relations.
Likewise, this ontology is working in conjunction with a Canonical Data Model (CDM), whose key role is to define a common vocabulary that will be used for communication among all the devices ensuring the syntactic interoperability of the RESPOND solution. The next figure showcases RESPOND’s ontology and CDM approach to integrate data coming from different sources.
Afterwards, this semantically annotated data is stored in an RDF Store, where it will remain accessible to be used by different applications or services. SPARQL (SPARQL Protocol and RDF Query Language) is the most commonly used query language to query the information available in an RDF Store, as it provides a set of query functionalities (e.g. join, sort and aggregate), together with graph traversal syntax.
Among the services that may exploit this semantically annotated data, the RESPOND project is developing a set of analytic services. These services include the demand forecasting at house and neighbourhood levels, which may derive in specific DR recommendations. Furthermore, this semantically annotated data is also available to be used by the user interface.
All in all, it can be concluded that Semantic Technologies are an important actor of the RESPOND project, as not only do they enable the integration of heterogeneous data, but also contribute to the development of services and applications.