Poster Abstract: Towards NILM for Industrial Settings

Industry consumes a large share of the worldwide electricity consumption. Disaggregated information about electricity consumption enables better decision-making and feedback tools to optimize electricity consumption. In industrial settings electricity loads consist of a variety of equipment, which can be difficult to monitor individually due to cost restrictions. We believe that Non-Intrusive Load Monitoring (NILM) can ease the burden of such a monitoring infrastructure. This hypothesis has been preliminary evaluated by collecting a data set, from more than forty sensors, measuring power consumption for six months, at an industrial site. In this poster abstract we provide initial results for how industrial equipment challenge NILM algorithms. These results thereby open up for evaluating the use of NILM in industrial settings.