A Statistical Study of Failures in the 10 kV Distribution Network of the Moscow Region

Abstract

The dependence of distribution electrical network outages on various factors is statistically studied, and the network reliability is analyzed based on the emergency log entries. The emergency log data are analyzed, put in a systematic order, and checked for correctness and completeness. The faults are combined into groups, outliers are considered, and anomalies are excluded from the sample using a statistical and qualitative method. Based on the processed statistical information, a correlation analysis is performed with the formulation and testing of statistical hypotheses; the sample numerical characteristics are calculated, and a theoretical law for distribution of the number of outages in the network as a function of the power supply restoration time is selected. Conclusions about the influence of the time of day and climatic factors on the faults in various distribution network facilities in the Moscow region have been drawn; the least reliable network equipment items have been defined, and a probabilistic mathematical model describing the probability distribution function of the power supply restoration time has been developed.