What the Semantic Web Will Mean for Smart Grids
- Written by Iñaki Laresgoiti
A really smart grid will, among other things, allow network control strategies to be adjusted to a dynamically changing environment so as to meet the requirements of all stakeholders. The adoption of advanced techniques that are being developed in the semantic web initiative will make grids more intelligent by automatically adjusting the situation of the network to the connection and to the removal of resources available to participate in the operation of the network.
The role of demand-side management techniques is destined to grow as renewable sources of power are integrated into competitive power markets. Unless energy demand is made more flexible, the need of ancillary services to maintain the current levels of reliability and security of supply will grow sharply. Thus, recruitment of stakeholders able to compensate for the intermittency of renewables would seem to be the most promising way of making electrical systems secure, clean, cost-effective and efficient, all at the same time.
Therefore, the generally conceived smart grid vision for year 2020 and beyond builds upon a myriad of dispersed components—on both the generation and consumption sides— that will be able to provide services to the network operation through an efficient and intelligent integration of distributed resources with other network components. Virtual power plants and microgrids will be key to the realization of the vision.
As extensive use of information and communication technologies will be required, and smart meters depending on those technologies are being widely deployed, directives are being issued to facilitate the smooth interaction of power system components, IT and communications. The European Commission Smart Grid Taskforce's Mandate 490, to be implemented via CENELEC , the European Committee for Electrotechnical Standardization, is a notable example.
The Internet and Web-based applications represent a particularly promising approach to the dynamic and efficient integration of distributed resources. The notion of a semantic web, which the World Wide Web Consortium (W3C) defines as "a common framework that allows data to be shared and reused across application, enterprsise and community boundaries," is the objective of the S-TEN project—Intelligent Self-describing Technical and Environmental Networks—which is funded by the European Commission. Tim Berners-Lee, the inventor of the Web and the director of W3C, has defined the semantic web as interconnected data "that can be processed directly and indirectly by machines," without human intervention.
The idea of S-TEN is to show how semantic web technologies can be leveraged to provide decision support in complex and potentially continuously changing networks based upon the use of semantic web technologies. So-called “ontologies,” considered one of the most important advances in software development, are an important aspect of the semantic web. The data described by an ontology is interpreted as a set of “individuals” together with a set of "property assertions," which relate these individuals to each other, and a set of axioms, which place constraints on sets of individuals (called "classes") and the types of relationships permitted between them. These ontologies, which represent the knowledge of a domain, provide semantics that allow systems to infer additional information based on the data explicitly provided.
Within the S-TEN project, such ontologies are used to represent knowledge associated with power system devices, ranging from simple sensors that measure an electrical feature, such as voltage, in a specific feeder of the network, to complex installations, such as combined heat and power plants or wind turbines, so that they can enroll themselves at a central registry. Registration exposes their capabilities and interfaces to applications searching for a matching service. In addition, the semantics describing equipment's design information and measurement data can be used together with knowledge of the service that has been requested by decision support systems, such as energy flow monitoring systems, to detect any malfunctioning of the equipment and to trigger corresponding alarms or some other kind of action. For example, an operator might be notified of best practices in the given situation
In future intelligent grids with a lot of distributed resources, such as photovoltaic arrays, wind turbines, combined heat and power plants, and electrical vehicles, each device will have its own intelligence to be registered in the network automatically and without any human intervention; the published information will include the description of all relevant information required for correct use of the offered services. Such registration is handled by devices that use the standard languages that are being promoted by the W3C.
As it comes to represent a complete inventory of all devices in the network, the S-TEN registry opens the possibility to electricity stakeholders, such as network operators, suppliers, "balance responsible parties" or others, of making semantically meaningful and detailed queries about the services being looked for and of getting the responses from the registry that include not only the best match to the services requested, but also all the relevant information to access them.
Semantic technologies mean rich data representation to perform complex reasoning in an open space as opposed to industrial process control, which implies low-level data to take quick decisions in a closed space. The evolution of existing electricity networks towards the smart grid makes necessary the enhancement of existing control capabilities with other capabilities that allow reasoning about information collected from different and dynamically changing sources. This higher level of reasoning and data representation makes mandatory the use of powerful ontologies representation languages, such the Web Ontology Language (OWL), which has been used within the S-TEN project.