Consumption awareness for energy savings: NIALM algorithm efficiency evaluation

The worldwide adoption of smart meters that measure and communicate residential electricity consumption gives rise to the development of new energy efficiency services. Among these services, it is worth mentioning the ones related to energy saving. In this context it would be essential to know the energy daily profiles of the appliances of a public utility customer. These information can be obtained by means of Non Intrusive Appliance Load Monitoring (NIALM) algorithms. The efficiency of such algorithm are related non only to the number and type of appliances, but also to the quality of the data provide by the power meters. In this paper, we study how the use of disaggregation algorithms in conjunction with smart meters is cost-effective and scalable solution for energy efficiency systems. New efficiency definitions and load indices are also introduced to evaluate the performance of disaggregation algorithms.