PQONT: A domain ontology for electrical power quality

https://doi.org/10.1016/j.aei.2009.06.009Get rights and content

Abstract

Electrical power quality (PQ) data is a ubiquitous type of engineering data quantified as a set of PQ parameters, such as frequency, flicker, and harmonics, obtained from electricity transmission and distribution systems. For PQ problems detected as deviations from nominal power system behavior negatively affect the workings of governmental and industrial institutions, as well as the electronic equipment and the psychology of individual humans, several systems of varying complexities to measure and monitor PQ parameters have been developed in order to detect the problems and take the necessary countermeasures. Yet, the domain of PQ data still suffers from the lack of a common vocabulary to be shared by the involved parties. In this paper, a novel domain ontology for electrical PQ, called PQONT, is presented to fulfill this need. Relevant standards and regulations have been utilized extensively during the ontology building process. The proposed ontology also supports linguistic applications by providing a number of linguistic properties applicable to all of its concepts. This facility of the ontology is employed in a multilingual natural language based interface for PQ data collected through a nationwide PQ measurement system so that the data can be queried in a flexible way. The implementation details of the interface is presented with a couple of query examples to illustrate the contribution of the ontology to a real application. The future work includes the extension of PQONT to a more general power system ontology so that it can contribute to a wider range of applications in the power system domain.

Introduction

In several application domains, dedicated ontologies have been proposed as the need for a common vocabulary of concepts in these domains arises. A domain ontology is defined in [1] as a reusable vocabulary of concepts, relationships, and activities in a particular domain along with the theories and principles governing the domain. A domain ontology called Kumbang is described in [2] for modeling the variability in software product families. Kumbang synthesizes the previous effort on the variability of the software product families and its semantics is presented in detail [2]. In [3], an ontology for the trajectory simulation domain is proposed, where trajectory simulations are defined to be the computational tools to calculate the flight path and flight parameters of munitions. The authors emphasize the possible contribution of their ontology to the interoperability of related simulation software [3]. Sensor networks data is the domain of the ontology proposed in [4] basically due to the ubiquity of the wireless sensor network applications. An ontological model to enhance the business processes of the companies in the electrical power industry is presented in [5] where the authors point out that their ontological approach will help the decision-making processes of these companies. These studies are considered here just to convey the breadth of the domain ontologies that have been proposed so far, and interested readers may refer to [1] for more examples of domain ontologies.

In this paper, an ontology developed for electrical power quality domain, called PQONT, is presented.1 Electrical power quality (PQ) data is a ubiquitous type of engineering data and is usually quantified in terms of a set of PQ parameters obtained from measured data at electricity transmission and distribution systems. The domain of electrical PQ is mainly concerned with the determination of problems that result in the degradation of the quality of the electrical power transmitted to the customers through the electricity transmission and distribution systems. These problems may cause malfunctioning or even breakdown of expensive electronic equipment which lead to considerable economical loss. Moreover, particular PQ problems manifested on light sources in domestic places, such as voltage flicker above acceptable limits, are known to have negative psychological effects on people. These problems call for proper monitoring of the quality of the transmitted power from the power plants where it is generated to the utilization sites such as industrial institutions and houses of people. This monitoring process should also be accompanied with ways of finding the causes of the problems and taking the necessary countermeasures. These motivations make the PQ domain a fruitful research area. To begin with, the studies including [7], [8] propose large-scale power quality monitoring systems which measure PQ parameters at particular sampling rates and allow monitoring of the measured data. However, these systems have proprietary interpretations of the concepts of the PQ domain and they lack explicit conceptual data models or domain ontologies which hurt the interoperability of the proposed systems since the systems do not produce directly comparable outputs. There are several other studies that propose exchange formats for PQ data ([9], [10], [11]), but these formats are not related to the semantics of the data allowing only the efficient exchange of PQ related waveforms in certain binary formats. Influential studies like [12] describe some subset of concepts in the PQ domain such as the types of PQ variations that can cause PQ problems. The authors [12] aim to guide the PQ researchers on the interpretation of a subset of concepts in the domain but the concepts described do not possess an ontological structure in terms of coverage and hierarchical structure. Considering the relevant IEEE and IEC standards [13], [14], [15], [16] and regulations [17], they are outputs of careful studies on the PQ domain encompassing domain and operational knowledge but they are not aimed to be used as formal ontologies hence they do not commit to provide a conceptual and hierarchical structure. Therefore, to the best of the authors’ knowledge, no generic ontology development effort for the PQ domain has been reported in the literature. The following points motivated us to propose our PQ domain ontology, PQONT:

  • A prospective PQ ontology can act as a common vocabulary for the domain, which can be used as a representative of the domain at the initial development stages of monitoring/measurement systems. Conforming to the semantic structure in the ontology can facilitate direct comparison of the measurements of distinct systems.

  • A PQ ontology can serve as an external semantic resource for many domain-specific information system applications including information extraction/retrieval systems or natural language interfaces just as the more general conceptual ontology, WordNET [18], for the concepts in the English language does for many applications.

  • A PQ domain ontology can also be used as a compact information source for researchers in the domain with its descriptions of the concepts, their properties, and relations in a hierarchical structure.

During the ontology development process, relevant IEEE and IEC standards have been extensively utilized along with previous conceptual studies such as [12]. Moreover, experience gained from the development of two distinct PQ monitoring systems both of which aim to detect PQ problems in the Turkish Electricity Transmission System has also made a considerable contribution to the development of PQONT. The first of these systems is a mobile measurement system which has collected sample PQ data from predefined measurement points during a predefined time interval to reveal the overall PQ status of the transmission system [19]. A database architecture has also been developed for the collected data which enables the interested users to query the PQ data through a visual query interface. The second system is the successor of the mobile system and aims to monitor PQ status in an on-line fashion through the deployment of permanent PQ monitors to the various problematic points in the transmission system [21]. During the development phase of these two systems, considerable amount of domain and operational knowledge has been acquired which has aided in the development of PQONT. PQONT has been created using the ontology development tool, Protégé [22] due to its extensibility and ease of interoperability with other applications. PQONT also supports linguistic applications through several properties of its concepts and thereby it has been employed in the implementation of a multilingual natural language based interface. Hence, the main contributions of the current study can be summarized as follows:

  • PQONT is the first explicit ontology developed for the domain of electrical PQ. It can contribute to the implementation and integration of PQ measurement and monitoring systems as well as other information-based systems acting as a generic vocabulary for the PQ domain.

  • PQONT provides linguistic support through a number of linguistic properties and it can be extended to support any language by adding the relevant linguistic properties for the language under consideration. This capability of PQONT has been demonstrated with the development of a genuine multilingual natural language based interface currently supporting two languages, namely English and Turkish, for PQONT currently contains the corresponding linguistic properties in these two languages.

From the perspective of the classical generic task approach for knowledge-based systems as proposed in [23] which points out that the systems should be made up of generic building blocks targeting at specific problems, the main tasks in the PQ domain that will benefit from PQONT include the following ones:

  • The design and integration tasks of especially PQ monitoring/measurement systems since conforming to PQONT at these phases will reduce the interoperability costs.

  • The task of mapping user queries provided in natural language into valid query expressions during PQ data querying through the natural language interface which employs the linguistic properties included in PQONT as will be clarified in Section 4. This task can be considered as an instance of the generic task of recognition.

The rest of the paper is organized as follows: In Section 2, an overview of electrical PQ parameters is provided. The details of PQONT are presented in Section 3. Section 4 is devoted to the description of the multilingual natural language based interface for PQ data utilizing PQONT. Finally conclusions with future work are presented in Section 5.

Section snippets

Electrical power quality parameters

Electrical PQ is measured at electricity transmission and distribution systems as a set of PQ parameters. In Fig. 1, a generic electricity generation, transmission, and distribution system is presented. At the generation site, there exist plants that generate electrical power using various sources including water, oil, and gas. The transmission system is responsible for delivering the power to the distribution companies. Finally, the distribution system receives the power from the transmission

PQONT: the domain ontology for electrical power quality

There are various standards such as [13], [14], [15], [16] and national regulations such as [17] concerning electrical power quality (PQ). There are also conceptual studies such as [12] that describe some important concepts in the PQ domain. Nonetheless, there exists no universal ontology for the domain that can not only act as a common vocabulary for the PQ concepts but also aid to the interoperability of the PQ applications. In order to fulfill this need, a domain ontology, called PQONT, is

A multilingual natural language based interface for PQ data

The storage of PQ data on dedicated databases and its processing to extract useful information is a relatively new application for the PQ domain. An explicit database model and a database architecture based on the model for PQ data are developed4 for the nationwide PQ data collected using the mobile measurement system presented in [19]. The system architecture is presented in Fig. 6.

The first component of the database

Conclusion

In this paper, a novel ontology for electrical power quality (PQ) domain, called PQONT, is presented. Although there are various studies on PQ data ranging from those on measurement and monitoring systems to the ones on data management issues involved, no explicit ontology development effort has been reported in the literature. The proposed ontology will serve as a shared vocabulary for the involved parties in the domain as well as a basic formal description and classification resource for the

Acknowledgements

This research and technology development work is carried out as a subproject of the National Power Quality Project of Turkey (Project No. 105G129, http://www.guckalitesi.gen.tr). Authors would like to thank the Public Research Support Group (KAMAG) of the Scientific and Technological Research Council of Turkey (TÜBİTAK) for full financial support of the project.

References (27)

  • T. Asikainen et al.

    Kumbang: a domain ontology for modelling the variability in software product families

    Advanced Engineering Informatics

    (2007)
  • A. Gómez-Pérez et al.

    Ontological Engineering

    (2004)
  • U. Durak, H. Oğuztüzün, S.K. İder, An ontology for trajectory simulation, in: Proc. of the Winter Simulation...
  • M. Eid, R. Liscano, A. El-Saddik, A universal ontology for sensor networks data, in: Proc. of IEEE International...
  • R.F. Navarro-Hernandez, A. Perez-Soltero, G. Sanchez-Schmitz, M. Barcelo-Valenzuela, An ontological model to support...
  • D. Küçük, Ö. Salor, T. İnan, I. Çadırcı, Building an ontology for flexible power quality querying, in: Proc. of the...
  • L. Cristaldi et al.

    A distributed system for electric power quality measurement

    IEEE Transactions on Instrumentation and Measurement

    (2002)
  • D. Divan et al.

    A grid information resource for nationwide real-time power monitoring

    IEEE Transactions on Industry Applications

    (2004)
  • S. Chen, X. Wang, Power quality XML markup language for enhancing the sharing of power quality data, In: Proc. of the...
  • W.W. Dabbs, D.D. Sabin, Employing extensible markup language (XML) in IEEE Std 1159.3-2002 PQDIF, in: Proc. of the IEEE...
  • E.W. Gunther, On creating a new format for power quality and quantity data interchange, in: Proc. of the IEEE 2005/2006...
  • C.J. Melhorn et al.

    Interpretation and analysis of power quality measurements

    IEEE Transactions on Industry Applications

    (1995)
  • IEC 61000-4-15, Testing and Measurement Techniques – Flickermeter – Functional and Design Specifications,...
  • Cited by (25)

    • A high-level electrical energy ontology with weighted attributes

      2015, Advanced Engineering Informatics
      Citation Excerpt :

      The details of the building process of the high-level electrical energy ontology, which is linked to FPQONT and FWONT, are presented in the next subsection and a summary of the main features of the electrical energy ontology is given in the last subsection. The text-related attributes of PQONT and WONT, label and synonymSet, can seamlessly be utilized to facilitate domain-specific applications like the natural language interface to the electrical PQ database, which relies on PQONT to determine the domain terms within the user queries, as described in [7]. Yet, in relatively more open-domain applications like document/text classification systems, the homonymy of the values of these attributes to other words may cause problems.

    • Semi-automatic construction of a domain ontology for wind energy using Wikipedia articles

      2014, Renewable Energy
      Citation Excerpt :

      For the domain of meteorological data, the ontology covered in Ref. [22], which is an ontology for atmospheric science, can be integrated into the wind energy ontology. For the domain of electrical power quality, PQONT domain ontology [9] has been proposed and it can readily be integrated into the wind power ontology as it is publicly available as an OWL file. Another plausible option is to extend the proposed ontology to make it a larger renewable energy ontology, considering the semantics of other renewable energy types like solar and biomass.

    • An ontology-driven framework towards building enterprise semantic information layer

      2013, Advanced Engineering Informatics
      Citation Excerpt :

      Respecting to the three classifications introduced in the introduction Section: (1) conceptualizing knowledge, (2) supporting design annotation and enrichment, and (3) serving as mediation media, some applications in the domain of design and engineering are investigated. Knowledge representation is for conceptualizing a certain domain, for example, ONTO-PDM [13] represented a product model in manufacturing environment, and [14] presented a domain ontology for electrical power quality (PQ) called PQONT. Ontology ONTO-PDM aims to facilitate the interoperability among software applications during the physical product lifecycle.

    • Classification of power quality events - A review

      2012, International Journal of Electrical Power and Energy Systems
      Citation Excerpt :

      In many countries, high-tech manufacturers often concentrate in industry parks, therefore any PQ events in the utility grid can affect a large number of manufacturers [11,12]. Therefore, if these unwanted variations in the voltage and current signal are not mitigated properly [13], they can lead to failures or malfunctions of many sensitive loads connected to the same system, which may be very costly for the end users. To mitigate the PQ events, it is necessary to identify the events through PQ events detection and classification system so that accordingly mitigation action can be carried out.

    View all citing articles on Scopus
    View full text