Multiple regression models as identifiers of power system weak points

Multiple regression models designed to identify the weak points of a power system are presented. On the basis of the determined reliability indices (duration of loss of load DLOL, and demand-not-supplied DNS), and by taking into account the apparent power difference criterion SDC as a voltage collapse proximity index, one can define the mutual relationship between these indices and the states of the particular power system components. The regression coefficients of MRMs indicate participation of each power system component (generators, transmission lines, substations, etc.) with regard to DLOL, DNS and SDC. Such information helps take measures of increasing the power system reliability and security. Subject-related results obtained in the Bosnian utility are given.