An ontology for fault diagnosis in electrical networks

This paper presents an ontology to represent the electrical network from the point of view of diagnosis, i.e., an ontology of the elements of the network that are necessary for fault diagnosis. The notion of `ontology´, rooted in philosophy, is used in knowledge engineering to describe explicit specifications of conceptualizations, where a conceptualization is a set of definitions of elements in a domain. It is a useful notion in knowledge based systems development because it provides the means for describing explicitly the conceptualization behind the knowledge represented in a knowledge base. An ontology, in this case for fault diagnosis, is a combination of small-scale ontologies. After a detailed analysis of the network and its behaviour, five viewpoints were identified as relevant for diagnosis. We generated an ontology for each to characterize more accurately and clearly the elements involved in the problem. Each of them focuses on different relevant aspects of the network and has a compact body of knowledge and a clear meaning in the problem. They are identified as transport, control, events, alarms, and trips. Each is discussed by the authors in relation to fault diagnosis