Abstract
The idea of using smart meters is growing rapidly in modern societies. It is important to collect and process fine-grained data of the meters to effectively facilitate all stakeholders of the smart grid. The classical sampling mechanism is time-invariant. Therefore, it results in the acquisition, transmission, and processing of a large amount of superfluous data. This shortfall is resolved by using the event-driven sampling.It renders real-time data compression. Afterward, the novel adaptive rate techniques are used for data segmentation and features mining. The pertinent features regarding the appliances consumption patterns are afterward used for their identification. It is realized by employing the Naive Bayes classifier. Results confirm a 4.4 folds compression gain and the processing gain of the suggested solution while securing 91.9% classification precision.