Bondarenko O. Modelling and Analysis of Operation Reliability in Power System Subsystems under conditions of Cascading Failure Development

The PhD thesis solves the problems of developing new and improving existing mathematical models and methods for diagnosing the technical condition (determining the service life) of switching equipment of power systems, modeling subsystems of electric power systems (EPS) to assess the operational reliability in case of cascading failures under conditions of uncertainty of the initial information. In the first section the conditions of operation, features, and strategic goals of the current stage of reforming the domestic power industry are analyzed. The problems and prospects of the development of the power grid complex of Ukraine are discussed. It is determined that there is a persistent trend towards an increase in accidents and the severity of consequences caused primarily by failures of power and switching equipment. In the second section presents a characterization of the disturbance modes in the subsystem of the power system during the failures of individual elements. The main factors leading to an increase in the number of emergency situations in the power system are identified. The necessity of developing models for assessing the technical condition, resource, and failure probability of power equipment in the analysis of operational reliability is justified. A method for constructing a comprehensive mathematical model for assessing the technical condition of high-voltage switches is proposed. A hierarchical structural scheme of a fuzzy logical inference for obtaining an aggregated assessment of the technical condition of a SF6 circuit breaker is developed. In the third section, complex linguistic mathematical models of a SF6 circuit breaker are proposed to determine the resource and failure probability. These models are based on the use of information on the technical condition parameters obtained in Online or Online and Offline modes. Fuzzy logic is suggested for combining the results of different measurement methods in Online and Offline monitoring modes to correctly assess the technical condition. The solution to the task of a comprehensive assessment of the technical condition involves creating fuzzy expert subsystems for the main functional units. The central element is a rule base and a mechanism for fuzzy inference. The proposed linguistic mathematical models of high-voltage switches allow determining failure probabilities, evaluating the risk of a deviation from normal mode, and determining the indicators of operational reliability of power systems. In the fourth section discusses methods and mathematical models for analyzing the risk in the power system subsystem under conditions of cascading accident development. Based on a deep analysis of the causes of event chains that significantly influence the occurrence and development of emergency situations in power systems, a generalized scheme for the development of an emergency in a power system is constructed. Taking into account the probabilistic nature of failures of the main power and switching equipment, as well as errors of dispatchers and operational personnel, the requirements for mathematical models for assessing the operational reliability of power systems in a cascade scenario of accidents are determined. Considering the significant level of uncertainty in information about the probability of states and the severity of consequences during the development of a cascade accident, a comprehensive fuzzy model for assessing the risk of developing cascade failures in the power system subsystem is developed. In the fifth section the task of modeling and studying the operational reliability of power system subsystems during failures and decommissioning of switching equipment and the cascade development of failures is addressed. Several types of models for determining the resource and failure probability of switches are investigated. A calculation and analysis of the results are carried out for making an optimal decision on decommissioning switches based on criteria using the Pareto optimization method and probabilistic-statistical modeling to assess the risk of an emergency situation in the power system by the Monte Carlo method.