“Detección de fallas en sistemas de distribución a partir de señales con pérdida de información por medio de metodologías de reconstrucción de señales dispersas”

Abstract

The imperative need to detect faults in distribution systems is still a challenge for distribution companies around the world, however, the existence of better communication and storage technologies could facilitate the aforementioned task, on the other hand, storing large quantities of data can mean a problem in systems where both physical and virtual space or in the cloud are limited. To deal with this problem, the need for a method that allows to acquire smaller amounts of data is proposed and then by means of signal reconstruction techniques, the collected information can be easily reconstructed and those reconstructed signals can be subjected to various analysis techniques that in the end they provide information on the existence of faults in addition to characterizing them. The present investigation works in the reconstruction of dispersed signals and in a method for the detection of faults in addition to characterizing them, at the same time it executes a comparative analysis against the diverse amount of data used for the aforementioned task. Finally, the document validates the use of the proposed methodology using real signals from a local distribution system.