Adopting the IEC Common Information Model to Enable Smart Grid Interoperability and Knowledge Representation Processes:The IEC Common Information Model (CIM)

The IEC Common Information Model (CIM)

The significance of the CIM standards relates to their function as a scalable and extensible semantic model for power systems. An authoritative description of its design and class composition is given in the associated IEC Standards (IEC 61970- 301, IEC 61968, and IEC 62325) and it is further described in [5961]. Miscon- ceptions about the CIM in terms of its use in database design and the ‘‘CIM compliancy’’ of technology interfaces are addressed in [62]. The structure of the CIM is designed to be flexible. It is object-oriented and presented as a Unified Modelling Language (UML) class model. Flexibility of the model derives from its properties of extensibility and scalability. Extensibility applies when new objects not available within the standard set are needed, they can be added, underlining the open nature of standard model. If these additions are considered of general use and subject to subsequent interoperability testing, they can become inducted into the internationally standardized version [63].

Examples of IEC CIM extension to suit utility use cases are numerous and reflect business case evolution in managing the smart grid through use of Model Driven Architecture (MDA). Extensions to the CIM can be categorized for dif- ferent purposes, such as widening its domain scope into substation equipment representation [64] or High-Voltage Direct Current (HVDC) modeling [65], to extending its ability to represent dynamic models for contingency analysis [66]. As it is canonical in its design, it is possible to integrate new ‘‘packages’’ of UML classes with dependency to the Core package as the scope of use cases for information exchanges widens. Nielsen and Neumann give a good overview of the processes associated with CIM extension management in [67]. An important recommendation from consensually accepted definitions of smart grid standards identified in [53] featured extension of the SIA to accommodate DERs. With respect to future smart grid operational requirements, this recommendation was responded to in [68] with a proposed design for an energy storage extension comprising a package of classes addressing control of grid-scale energy storage technologies. The CIM is also being used as the design basis for a variety of new model-driven applications including state estimation [69], wide area measurement [70], and secondary equipment management [71].

The CIM is designed to be scalable, such that if a subset (or profile) of the standard reference classes are sufficient to model a given use case in a particular context then the rest of the reference metamodel can be ignored. Well-established profiles such as the CPSM and CDPSM have already been mentioned but the tendency to profiling for reusable functionality within the exchange of network models has become more common. The second edition of the European Network of Transmission System Operators for Electricity (ENTSOE) profile version 2.0, which was based upon CIM release 15, is an example of a combination of a bundle of standardized CIM profiles, each referring to specific functionality, including:

• Geographical profile, IEC 61968-13;

• Equipment profile, IEC 61970-452;

• Diagram layout profile, IEC 61970-453;

• State variables profile, IEC 61970-456;

• Topology profile, IEC 61970-456;

• Dynamics profile, IEC 61970-457.

The relationships between CIM UML classes are structured to provide a standardized object-oriented modeling architecture. It is a canonical taxonomy in the form of packages of UML class diagrams referring to the components of power utility networks with functional definitions and measurement types to a high degree of granularity. Wang and Van Ausdall give an overview of how business data semantics are represented in the CIM and propose some rules to clarify the UML modeling concepts used [72]. They describe how an XML namespace defines the scope of a class name and observe how a CIM class name (and therefore the concept represented by that CIM class) must be unique within the CIM XML namespace to maintain the integrity of the CIM logical model. This raises the distinction between the CIM as a static logical model, a standard con- ceptual representation of smart grid components, and the instantiation of CIM objects in models created by PSA CIM adaptors to represent their functional data models.

Power system applications use the CIM as a reference logic when processing CIM models for export and import. CIM metadata files communicated between PSAs vary in size depending on the scope of the modeled network (for example, a transmission system with complex topology) as well as the detail of the CIM representation of network parameters being communicated. With the most detailed representations of complex networks made up of millions of CIM objects forming multi-Gigabyte sized files, concerns over the amount of data, and the capacity to handle it within the smart grid environment may arise. This topic has been acknowledged and addressed by McMorran in [73] in which a number of strategies are discussed for reducing the size of, and handling, communicated CIM files. The principal strategies for handling large CIM representations of power system net- works include communication of layered representations of a network constrained to CIM profiles (see profiles above); the use of difference models (see IEC 61970-

552) that only update the status of larger parent models as changes to them occur; the use of compression technologies such as the ZIP file format that can perform better than 20:1 compression on CIM RDF XML. It is unlikely therefore with RDF forming the backbone of communicated CIM files that any great stress will be placed on the data communications and storage capabilities within the smart grid environment.

The semantic definitions and logical integrity of the exchanged model depends on the CIM standards but its ‘‘physical’’ integrity or connectivity depends upon a system of object identification provided by RDF. RDF links objects together by means of a triple, defining a subject in relation to an object using a predicate. The predicate as a system of address is used to form the identity description of the object and is generated within the CIM adaptor of the PSA when processing a CIM model. An instantiated model of CIM objects must conform to the logic and

semantic definitions of the standard CIM static model but will only use a portion, or profile, of its set of CIM classes to represent the real network. If each inter- operating PSA places its instantiated CIM objects within the same namespace, such as ‘‘xmlns:CIM,’’ then the opportunity for object identity collisions will arise when these models are shared [74]. This is because the namespace defines the scope of validity for an object identity just as it does for the semantic descriptions of the object. Identity collisions therefore are a vexing problem currently chal- lenging smart grid PSA interoperability.

The CIM as Ontology for the Electrical Power Domain

If we consider a model as ‘‘an abstraction of reality according to a certain con- ceptualisation’’ [75], then these standardized models, as meta-conceptualizations representing PSA data models, support the view of the CIM as a metamodel in accordance with [76, 77]. The canonical nature of the CIM in giving rise to a range of submodels (profiles) that describe specific context-constrained applications enable it to also be described, in terms of a ‘‘model of models’’ which concurs with the Object Management Group (OMG) definition of a metamodel [78].

Harmonization with other existing information models, such as the IEC 61850

substation automation standard, to widen the integrated semantic standards framework supporting smart grid interoperability is seen as a priority. Gruber defines ontology as a ‘‘specification of a representational vocabulary for a shared domain of discourse—definitions of classes, relations, functions and other objects’’ [79]. As the scope of the CIM extends, placed at the heart of a harmonized federation of standards, it conforms to Gruber’s definition of ontology for the smart grid domain. In this sense ontology supports the description of our knowl- edge about a domain, linking the IEC CIM to knowledge representation of the smart grid. This proposal is fundamental to the capacity of the CIM within the smart grid domain for knowledge representation and sharing. Chandrasekaran et al. argue it is not the representational vocabulary of the domain that defines the ontology as much as the conceptualizations that the vocabulary is intended to capture [80]. Careful analysis of the objects and their relationships within the domain is required to create the vocabulary and conceptualizations necessary for true representation of the domain reality and explains why CIM development is marked by much debate amongst domain experts as well as the importance of interoperability testing. For, as Uslar et al. indicate in [81], the strength of the CIM as a domain ontology not only depends on the expertise of the domain experts building it, but also extending its application to link control center ICT with field- automated devices while further developing the SIA.

Harmonization of the CIM with Other Standards

Regarding the link between the CIM and field devices, Santodomingo et al. [43] discuss the harmonization of the CIM with IEC 61850 (substation control lan- guage) using an ontology matching approach that required the use of Web Ontology Language (OWL) to represent semantic correspondences between the two standards. Their methodology was based on a top-down application of service descriptions that were used to annotate CIM metadata mentioned in [36]. The CIM and IEC 61850 ontologies supported a layered framework created to bridge the semantic definitions of their classes and attributes and the relationships of these entities. In this way the harmonization of these two standards, designed from different origins and for different purposes but now increasingly required to interoperate to develop smart grid functionality, is being established.

In another initiative, linking the CIM to IEC 60870 for high-voltage meter control and management is described [82]. The semantic alignment of these two standards is seen as part of the development of the Spanish smart grid. Mapping of the classes from the IEC 60870 protocol to the CIM was reported as straightfor- ward and described in the sense of aligning one ‘‘service’’ to another. The sense of model classes representing services is another indication of the way the CIM lends itself to SOA. What is more, with the application of ‘Simple Protocol and RDF Query Language’ (SPARQL) the opportunity to interrogate RDF databases annotated with metadata makes possible the benefits of the Semantic Web para- digm. SPARQL is designed to seek out query matches with RDF triples for data stored in an RDF format such as CIM RDF XML. In this case the use of multiple namespaces, as metadata annotation of the meter data captured in CIM RDF XML, enabled the machine-to-machine (M2M) access required by the query. This methodology presents another example of how a layered architecture builds interoperability between the source of data and an end use. Whereas the use of Web Ontology Language (OWL) as a layer will focus on the resource description logic, SPARQL will focus on the knowledge representation of the RDF triple.

Related posts:

The Efficient Use of Energy:Load Factor and Power Factor.
Switchgear:High-voltage (transmission) switchgear.
Transformers:Standards.
Direct-Current Voltage Testing of Electrical Equipment:Evaluation of Test Data Readings
Cables and Accessories:Latest Trends in Cable Condition Monitoring and Aging Assessment
Testing and Commissioning of Protective Relays and Instrument Transformers:Types of Relay Tests
Electrical Power System Grounding and Ground Resistance Measurements:Ground Resistance Measurements
Starting and generating systems:Charging systems
FUEL CELLS:FUEL CELL EFFICIENCY
WIND POWER
General Overview:Electricity Generation
The Current Situation and Perspectives on the Use of Solar Energy for Electricity Generation:Concent...
The Current Situation and Perspectives on the Use of Geothermal Energy for Electricity Generation:Ge...
The Current Situation and Perspectives on the Use of Nuclear Energy for Electricity Generation:Nucle...
Energy Hub Management with Intermittent Wind Power:Discussion

Leave a comment

Your email address will not be published. Required fields are marked *