Intrusive and Non-Intrusive Load Monitoring (A Survey)

Marco Danilo Burbano Acuña

Abstract


There is not discussion about the need of energyconservation, it is well known that energy resources are limitedmoreover the global energy demands will double by the end of2030, which certainly will bring implications on theenvironment and hence to all of us.Non-Intrusive load monitoring (NILM) is the process ofrecognize electrical devices and its energy consumption basedon whole home electric signals, where this aggregated load datais acquired from a single point of measurement outside thehousehold. The aim of this approach is to get optimal energyconsumption and avoid energy wastage. Intrusive loadmonitoring (ILM) is the process of identify and locate singledevices through the use of sensing systems to support control,monitor and intervention of such devices. The aim of thisapproach is to offer a base for the development of importantapplications for remote and automatic intervention of energyconsumption inside buildings and homes as well. For generalpurposes this paper states a general framework of NILM andILM approaches.Appliance discerns can be tackled using approaches fromdata mining and machine learning, finding out the techniquesthat fit the best this requirements, is a key factor for achievingfeasible and suitable appliance load monitoring solutions. Thispaper presents common and interesting methods used.Privacy concerns have been one of the bigger obstacles forimplementing a widespread adoption of these solutions; despitethis fact, developed countries like those inside the EU and theUK have established a deadline for the implementation ofsmart meters in the whole country, whereas USA governmentstill struggles with the acceptance of this solution by itscitizens.The implementation of security over these approachesalong with fine-grained energy monitoring would lead to abetter public agreement of these solutions and hence a fasteradoption of such approaches. This paper reveals a lack ofsecurity over these approaches with a real scenario.

Keywords


sensors; machine learning algorithms, smart meter

Full Text: PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.