Design and Implementation of a Short Circuit Detection System Using Data Stream and Semantic Web Techniques

The present work focuses on the design of a system for the detection of short circuits using semantic web techniques and data stream. An architecture for the system is designed, as well as the communication model to be used. The different processes through which the data pass from the moment they are generated in the sensors until they are visualized by the user are described. The machine learning algorithm (K-means) is applied to classify the streams. The results are mapped to the existing graph in the triplestore, and the results are displayed on a geographic map. This chapter focuses on the design of a system for the detection of short circuits using semantic web techniques and data stream. Finding short circuits on transmission lines and poles has been a matter of concern for network operators, as lines are known to be characterized by immense lengths and large numbers of poles. In Cuba there are many failures in the national electrical system, mainly caused by short circuits. Streaming annotation ontology is one of the representatives incorporating IoT data stream, based studies: TimeLine, PROV-O, Social Security Number, and event ontology. IoT and semantic web technologies have evolved to create data models and systems that support the development of top-level applications in the field of computing and telecommunications. The design of the short circuit detection system has all the elements used in IoT technologies for semantic web systems because it combines grouping models and incremental reasoning to facilitate decision making.