Fuzzy c-means clustering for steady state events classification of electrical signals

Improving the efficiency of the electricity usage is an important concern these days relaying on monitoring techniques mainly divided into two categories: the intrusive monitoring that is focused on measure the power consumption of each electrical load individually and the non-intrusive monitoring which measure the power consumption from a centralized point and determine which electrical loads are presented. Electrical signatures are the main components of a non-intrusive monitoring system; these signatures are presented on the transient and steady state events on the signal. In order to classify these events according to its signatures, it is necessary to detect when and where these events occur. This work presents a method to detect and classify steady state events using the fuzzy c-means clustering algorithm; the proposed method is based on a frequency domain analysis, more specifically the analysis of the fundamental frequency and the first odd harmonic in order to improve the computational cost of the event classification algorithm.