The results of testing the algorithms of the adaptive model of a synchronous generator using theoretical and real physical data are presented in this study. The adaptive model of a synchronous machine is an equations system, which describes both the static and transient operation of a generator. Parameters of the adaptive model are found using measurements of a generator’s operational parameters. The single-machine model was created in Matlab/Simulink software to test the theoretical data. This single-machine model consists of a synchronous generator, a step-up transformer, and a transmission line. The test model also includes models of the automatic voltage regulator and steam turbine governor. The real electrodynamic model was used to verify the adaptive model of a synchronous machine. It consisted of four synchronous generators, with values of power capacity of 5 kW and 15 kW. The data logger with a sampling rate of 57.8 kHz was developed and installed to measure the operating parameters of each generator. As a result of testing on both models, the following values were estimated: inertia moment, d-axis and q-axis reactance, and load angle. These values were compared with the reference values. The adaptive model of a synchronous machine can be used in systems of emergency control and assessment of generator state.

The problem of identifying and damping low-frequency oscillations in the electrical operation mode is becoming particularly relevant for modern electric power systems containing renewable energy sources, flexible control devices based on power electronics, and featuring reduced margins of electrical network transmission capacity. Modern studies aimed at development of methods for assessing the parameters of low-frequency oscillations are reviewed and analyzed. It is shown that the existing methods are insufficiently efficient due to a significant delay and low adaptability. The article proposes a method for fast estimation of the parameters of low-frequency oscillations with a delay equal to a quarter of the oscillation cycle. The proposed method was tested on synthetic and physical signals. A digital model of a power system consisting of four synchronous generators was used as a source of synthetic data. Physical data are records of transients occurring in a real power system. To determine the accuracy of parameters characterizing low-frequency oscillations of the electrical mode, the article proposes a technique based on the analysis of the difference between the original signal and the signal reconstructed from the calculated parameters. As a result, the error values for the estimation of low-frequency oscillations were obtained, which confirm the accuracy and efficiency of the technique when analyzing both synthetic data and data from real power systems. Задача выявления и демпфирования низкочастотных колебаний электрического режима становится особо актуальной для современных электроэнергетических систем с наличием возобновляемых источников энергии, гибких устройств управления на основе силовой электроники и сниженными значениями запаса пропускной способности электрической сети. Проведен обзор и анализ современных исследований, направленных на разработку методов оценки параметров низкочастотных колебаний. Выявлена недостаточная эффективность существующих методов, обусловленная значительной задержкой оценки и низкой адаптивностью. В статье предложен метод ускоренной оценки параметров низкочастотных колебаний с задержкой, равной четверти цикла колебаний. Апробация метода выполнена на синтетических и физических сигналах. В качестве источника синтетических данных использована цифровая модель энергосистемы, состоящей из четырёх синхронных генераторов. Физические данные представляют собой записи переходных процессов, протекающих в реальной энергосистеме. Определение точности оценки низкочастотных колебаний параметров электрического режима основано на анализе разности исходного и восстановленного по вычисленным параметрам сигналов. В результате получены значения ошибки оценки низкочастотных колебаний, подтверждающие точность и эффективность метода при анализе как синтетических, так и данных из реальных энергосистем