Attribute selection for fault location recognition in transmission lines

After a severe disturbance due to an insulation failure in a transmission line, the precise fault location is a critical problem for the maintenance crew. In order to avoid further economical and social costs, fault diagnosis has to be performed as soon as possible. Fault diagnosis has been a major area of investigation among power system problems and intelligent system applications. Several approaches have been proposed for solving this problem. This paper advocates the application of Support Vector Machines for mapping the relationship between electrical signals and fault locations in transmission lines. The significance of voltages and currents is analyzed using steady-state and electromagnetic transient information, i.e., two ways of feeding the fault location models are compared. The tests consider different operating and fault conditions, including different types of fault, a variety of fault impedances, fault angles, line loading, equivalent system impedances, and fault locations.