k-means algorithm and mixture distributions for locating faults in power systems

Enhancement of power distribution system reliability requires of a considerable investment in studies and equipment, however, not all the utilities have the capability to spend time and money to assume it. Therefore, any strategy that allows the improvement of reliability should be reflected directly in the reduction of the duration and frequency interruption indexes (SAIFI and SAIDI).In this paper, an alternative solution to the problem of power service continuity associated to fault location is presented. A methodology of statistical nature based on finite mixtures is proposed. A statistical model is obtained from the extraction of the magnitude of the voltage sag registered during a fault event, along with the network parameters and topology. The objective is to offer an economic alternative of easy implementation for the development of strategies oriented to improve the reliability from the reduction of the restoration times in power distribution systems.In the application case for an application example in a power distribution system, the faulted zones were identified, having low error rates.