Diseño e implementación de un sistema de monitoreo de carga no intrusiva para el sector residencia

Abstract

The consumption of natural resources has increased exponentially since the industrial revolution, and the vast majority of these are not unlimited. This has led to the development of various technologies that prevent us from avoiding inefficient spending of resources. Within these technologies, we have non-intrusive load monitoring (NILM). Basically, this is based on the analysis of the energy consumption characteristics of the different household devices. This analysis allows us to disaggregate the energy consumption of a home, to take measures of the consumption characteristics in its connection, with non- intrusive methods. To this end, this thesis has focused on using artificial neural networks to determine if a home device is turned on or not. This is achieved by training the red neuronal with previous measurements of the consumption patterns of the different devices to identify their state. And so that the user can have a simple and mobile feedback of all the extracted information, an Android application has been developed that indicates both the status of the device and an approximate of current consumption and active power.