Control of electric energy consumption in steel industry using knowledge based techniques

Pekka Koponen, Raimo Viherma, Timo Rämö, Paavo Uronen

    Research output: Contribution to journalArticle in a proceedings journalScientificpeer-review

    Abstract

    Highly varying electric power consumption is charasteristic to steel and it is expensive to meet these peak loads. Is seems possible to predict the short term power demand and to use these predictions to control the plants so that expensive simultaneous peak loads can be avoided. In order to test this idea we built a prototype system. There are several reasons for applying knowledge engineering to this problem. Object oriented programming is powerful in the development and maintenance of the system and its prediction and structure models. The models of the steel making processess as well as the decision making knowledge are often heuristic. Also prototypes are needed in order the get the design feedback from the end users. In this paper the problem and the prototype are described. The prototype mostly consists of Lisp-programming structured with frames. we have used rules only in one knowledge base. Good user interface development tools turned out to be essential for the prototype building. The purpose of the prototype is to provide a basis for the definition of the real system. The prototype demonstrates a proposed solution but leaves some questions unanswered. Nevertheless the prototype shows how power can be predicted and that these predictions are important in controlling the peak loads
    Original languageEnglish
    Pages (from-to)31-37
    JournalIFAC Proceedings Volumes
    Volume25
    Issue number17
    DOIs
    Publication statusPublished - 1992
    MoE publication typeA4 Article in a conference publication

    Keywords

    • Steel industry
    • metals production
    • power management
    • prediction
    • knowledge engineering

    Fingerprint

    Dive into the research topics of 'Control of electric energy consumption in steel industry using knowledge based techniques'. Together they form a unique fingerprint.

    Cite this