Research on the model of association rule based on alarm ontology in substation

This paper is to further explore the information of alarm in substation, and to improve the efficiency and accuracy of association rules. It comes up with a mining method of association rule based on alarm ontology, and builds up a model of it. This method divides the alarm into 6 top concept and 13 bottom concept on the basis of the construction knowledge of ontology, the information generalization and preparation of the data dictionary, to set foundation for the improved Apriori algorithm and also for the feedback mechanism in the model. Compared to the existing improved algorithms, this method can not only improve operational efficiency, but also reduce redundancy and useless conclusion. The validity of this method and model is proved by the efficient generation of association rules and assertions rules.