Stochastic assessments of voltage sags due to short circuits in electrical networks

Abstract

Abstract : Short duration voltage sags due to short circuits are generally considered as a serious power quality/electromagnetic compatibility problem. The main concern at the moment is the tripping of adjustable-speed drives. They can be extremely sensitive to voltage sags and their size makes mitigation methods expensive in investment, maintenance and additional losses. To decide about the suitability of mitigation methods, knowledge is needed about the expected number of sags as a function of magnitude, duration and other characteristics. This knowledge can be obtained through the use of power quality monitors or by means of stochastic prediction methods. Theoretical studies as well as simple counting of events, have shown that a large fraction of the voltage sags experienced by a sensitive load originate in the transmission and sub- transmission systems. Therefore, special emphasis is needed on monitoring and stochastic assessment studies in transmission and sub-transmission systems. Monitoring has its limitations: the monitoring period can be several years if accurate results are required; the results of system changes are difficult to assess; extrapolation to sites not monitored could give unreliable results. Stochastic prediction techniques do not have these limitations: they give an accurate result immediately. The main aim of this research has been the development of models and techniques for stochastic prediction of the number of voltage sags in the electrical network. In this thesis, a so-called "method of fault positions" is used to estimate the expected number of voltage sags as a function of their characteristics (magnitude, duration). This method proceeds as follows: for a large number of fault positions in the system under study, the magnitude and duration of the voltage sags at user positions are calculated by using circuit theory techniques. An expected number of fault occurrences for this fault position is determined by using basic power system reliability techniques. The results of all the fault positions are added to obtain the expected number of sags. The method has been applied to a large transmission system and a distribution system, for which impedance data were obtained and failure rates were estimated. The results of the calculations have been presented in a number of ways: voltage sag tables showing the expected number of sags per year; maps showing the sag at one substation due to faults at many other locations; maps showing the sags at many substations due to a fault at a specific location, and maps showing the expected number of sags per year at all substations. The results are analysed and conclusions are drawn. The possible uses of these techniques are shown by applying them to the IEEE Reliability Test System. In the Reliability Test System, sensitivity analyses are carried out in order to analyse the influence of the system operation on the number of sags. Also, three-phase-faults, single-phase-to-ground and phase-to-phase faults are included. Apart from magnitude and duration, the jump in voltage phase-angle, and the three-phase imbalance are presented in the results.