Unsupervised event extraction within substations using rough classification

Ching Lai Hor, Peter A. Crossley

    Research output: Contribution to journalArticlepeer-review


    Microprocessor technology, broader bandwidth communications, and cheaper storage medium have greatly improved the capability to process, transmit, and store the large quantities of data available in a substation. Intelligent electronic devices (IEDs) normally acquire some of these data as raw facts, which then need to be interpreted in order to extract the useful information that engineers and operators require. Human interpretation is becoming increasingly impractical and the effect can hamper, or even prevent, an operator responding correctly to an emergency. This paper explains how a rough classification technique enhances the capabilities of substation informatics and provides valuable insight into the information contained in a substation dataset. This paper emphasizes postfault analysis of the protection and breaker responses in a substation. It is designed to help the operator understand overwhelming alarm messages or longer term to help engineers analyze what went wrong. The formulated methodology is generic and applicable to any type of transmission and distribution substation. © 2006 IEEE.
    Original languageEnglish
    Pages (from-to)1809-1816
    Number of pages7
    JournalIEEE Transactions on Power Delivery
    Issue number4
    Publication statusPublished - Oct 2006


    • Discernibility matrix
    • Intelligent electronic devices (IEDs)
    • Knowledge extraction
    • Rough sets
    • Unsupervised event extraction


    Dive into the research topics of 'Unsupervised event extraction within substations using rough classification'. Together they form a unique fingerprint.

    Cite this