Printer Friendly

Architectural issues in CIM to relational model conversion.

Abstract: The deregulation of the electrical utility industry has made it necessary to develop certain standards which would allow various participants in the industry to exchange electronic information efficiently. A standard which addresses this need is the Common Information Model (CIM) developed by the IEC. This paper will propose a solution which would allow companies using relational models of power systems to effortlessly exchange data with other companies involved in the industry by exporting/importing model data to CIM compliant RDF/XML documents. The paper will demonstrate the research results, which included exporting a model stored in a relational database to RDF/XML and importing the same data to the relational database.

Key words: Common Information Model (CIM), Distribution Management Systems (DMS), Unified Modelling Language (UML), Relational Database Management Systems (RBMS)


The deregulation of the electrical utility industry required certain standards to be developed for data exchange between various energy/distribution management systems (EMS/DMS). To address these needs, the IEC developed a family of standards named Common Information Model (CIM). These standards define an object model which should be implemented by all the parties of the electrical utility industry in order to allow efficient data interchange.

Distribution Management Systems (DMS) are software systems which allow modelling, simulation and control of power systems on the distribution level. This usually means hundreds of thousands of values in dynamically changing systems which are frequently reconfigured by switches.

On the other hand, Energy Management Systems focus on generation and transmission, usually deal with far less variables and the model changes infrequently (Neumann S., 2001).


Both EMS and DMS consist of hardware and software components. Different companies involved in power generation, transmission and distribution use different software solutions. In a deregulated marketplace these companies need an efficient and standardized method for electronic data exchange. The CIM was developed to address this need.

Most of the applications developed for DMS/EMS use relational models of power systems and relational databases for data storage. The conversion of the existing models and storage mechanisms to CIM compliant solutions is a tremendous undertaking.

An intermediate solution could be to retain relational models and storage and implement only a CIM compliant data exchange layer. This paper proposes one such solution which allows DMS/EMS applications to convert data from a relational model to CIM and vice versa.

RDF (Resource Description Framework) is widely adopted for data exchange in various systems. RDF is an open standard developed by the World Wide Web Consortium (W3C). RDF data can be saved as a special XML document.


The solution which will be proposed in this paper allows data interchange between various relational models (stored in relational databases) and CIM. It is generic and easy to modify and allows the accomodation of various relational models. The functions which the tool should implement are shown on the use case diagram in Figure 1.


The actor named DataExchanger is any part of a DMS/EMS which requires CIM data exchange. The model mentioned in the names of the two central use cases refers to the Common Information Model.

In order to allow generic data input and output, the various electrical objects defined by CIM are stored in generic containers as defined in Figure 2.


The type of a certain PowerSystemResource (PSR) is determined based on its CIMType. Connectivity objects are used for describing the topology of the network. These Connectivity objects are an aggregation of CIM's ConnectivityNode and Terminal types. The double association between PSRs and PSRContainers allows effortless implementation of a multilevel hierarchy of objects. For example, at the highest level reside Substations, which can contain PowerTransformers, Compensators and a plethora of other CIM objects. A container type which is quite often found inside a Substation is a Bay, which usually contains Breakers and protection equipment.


Data output is the process in which a relational model used by a DMS/EMS application is loaded from a relational database, converted to CIM and then saved in RDF/XML. In order to achieve this, data from the relational database has to be fetched in a specific way, to allow easy population of the generic CIM containers presented in Figure 2. This can be done by creating generic RDBMSTable objects which fetch data from specified database tables and populate a certain type of CIM objects. The discussion of these RDBMSTables spans beyond the scope of this document. Another important step is the conversion of the relational model to CIM. Some conversion steps can be done by only modifying SQL commands, but for certain models some parameters have to be specifically calculated.

It is very useful to have all the attributes of PSR objects stored in PSRAttributes types (see Figure 2. in which the note stresses that PSRAttributes are name/value pairs), as it makes easier and more generic both data import and export.

The output of data isn't finished by simply loading it into a memory based structure shown in Figure 2. For effortless data exchange the data has to be saved in a format which allows efficient business to business (B2B) integration. IEC recommends using RDF/XML for data exchange.

CIM data, if stored in the way proposed in Figure 2, can be easily saved to RDF/XML, by only creating configurable RDF writers, which based on their configurations save various CIM objects to specified XML document parts.


Data input is the process in which data is loaded from RDF/XML into CIM and eventually stored in a relational model. Two aspects of data input will be discussed:

1. input from RDF/XML into CIM

2. saving CIM data into a relational database (conversion from CIM to a relational model is included in this step)

The first step, the population of CIM from an RDF/XML document holding the snapshot of the electrical network can be done by far more easily than converting it to a relational model and saving the data to an RDBMS.

5.1. Input from RDF/XML

The data structure in an RDF/XML is very similar to the data storage mechanism introduced in Figure 2. Listing 1 shows a part of an RDF/XML document describing one substation object.
Listing 1. RDF/XML representation of a CIM Substation

<cim:Substation rdf:ID="Substation_359158">
rdf:resource="#PSRType_2" />

As we can see in the above code snippet the substation has a few attributes stored in various parts of the XML tag. These are read into the PSRAttributes of a PowerSystemResource. Each attribute is stored in either an XML node attribute (e.g. rdf:ID) or in XML node content (e.g.

5.2. Saving to a relational database

In contrast to the data in RDF/XML, which is similar in structure to the data storage structure proposed in Figure 2, the various relational models will probably differ from it considerably. These differences may include:

1. one CIM object mapped to multiple database tables

2. multiple CIM objects mapped to one database table

In order to accommodate different relational models, it is not advisable to write separate writers for every CIM object. Generic, configurable writers have to be written. Figure 3 shows a class diagram, which addresses this requirement.


The RDBFactory is a storage facility for various CIMWriters, which are configured for storing a certain class of CIM objects. Each CIMWriter can write to multiple tables in the relational database used for the storage of the relational model. The actual mapping of PSRAttribute names to database column names is done through RDBColumns objects.

Each of the classes from Figure 3 is generic and configurable enough to allow saving CIM data to almost any relational database structure. The configuration of these classes can be stored as an XML document.


The solution proposed in this paper has a real life benefit, as it eases the first step towards full CIM compliance for DMS/EMS application developers. Its generic nature allows it to be used for data exchange between various, proprietary models stored in relational databases and the open CIM. CIM in turn, if stored in RDF/XML, can be easily understood by various software systems.

The prototype of the solution was tested on the relational model developed by the DMS Group Ltd., Novi Sad, Serbia & Montenegro. The testing was successfully completed on a simple network with a couple of substations.

Further research could involve the full integration of the IEC-61968 family of standards which is more closely related to distribution management systems.


Wang X., Schulz N.N. & Neumann S. (2001), "CIM Extensions to Electrical Distribution and CIM XML for the IEEE Radial Test Feeders", IEEE CIM Extension proposal

S. Neumann, (2001) "CIM extensions for electrical distribution," Proceedings of the 2001 IEEE Power Engineering Society Winter Meeting, pp. 904-907.

Booch G., Rumbaugh J. & Jacobson Ivar (2000), "UML--Vodic za korisnike", CET, ISBN 86-7991-111-9

Booch G., Jacobson I. & Rumbaugh J. (1999), The Unified Modeling Language User Guide, Addison-Wesley

Common Information Model (CIM): CIM 10 Version (2001), EPRI, Palo Alto, CA, 1001976

Draft IEC 61970 Series Standards, Energy Management System--Application Program Interface (EMS-API), Part 501, CIM RDF Schema, revision 2.

RDF Primer, W3C Recommendation. [Online]. Available:, accessed 2005-05-15
COPYRIGHT 2005 DAAAM International Vienna
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lendak, I.; Erdeljan, A.; Perlic, Z.
Publication:Annals of DAAAM & Proceedings
Geographic Code:1USA
Date:Jan 1, 2005
Previous Article:Reactor controlled switching system--factory and field testing and service experience.
Next Article:Not straight line enhancement and detection.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters