On the possibility of preventive assessments of damage rate of electrical networks (on the example of PJSC Rosseti Lenenergo)

Abstract

The article is a continuation of the study of operational reliability level of electric networks (EN) of PJSC Rosseti Lenenergo in 2014 – 2021. The classification of causes of damage in the EN and identification of predominant among them in the Company's EN during the operation period are considered in detail. The purpose of the article is defined as the study of cause-and-effect relationships that determine the accident rate, for which a number of tasks are formulated to be considered. The input data are the results of annual reports on the number of failures, broken down by months from 2014 to 2021 in the Leningrad region (LO) and St. Petersburg (SP) electrical networks. As research methods, numerical methods for evaluating the study results and the MATLAB graphic editor technology were used to visualize the analysis results of the studied indicators. Methods of mathematical statistics and artificial neural networks were used as preventive damage assessment methods.To approximate the series of emergency outages, programming algorithms developed by the authors were used, which enabled the visualization of the studied characteristics. It is established that the most appropriate solution is to use statistical methods based on the least squares method. With their application, it is possible to study trend-seasonal models that allow to assess the influence of the seasonal component on the number of failures of electric grid elements.As a result of the study, the main causes of failures of elements of electrical networks have been identified. For the networks of the LO, the main damage cause is the fall of trees (41%), for the SP networks — the impact of unauthorized persons and organizations (48%). Trend and trend-seasonal models have been constructed that allow obtaining short-term forecasts of emergency outages. At the same time, the assessment of the seasonal component reflects the dependence of the number of accidents on the month in which they occurred. For a more detailed analysis, it is planned to consider the influence of climatic parameters on the studied indicator.