THE ANALYSIS OF THE DYNAMICS OF CHANGE OF CRITERIA USED FOR INTERPRETATION OF DGA RESULTS, IN CORRECT HIGH-VOLTAGE TRANSFORMERS OF NON-GERMETIC EXECUTION

• Main Author: O. V. Shutenko
• Format: Article
• Language: English
• Published: Zaporozhye National Technical University 2018-02-01
• Series: Elektrotehnìka ta Elektroenergetika
• Subjects: 5 dissolved gas analysis; concentration of gases; slew rate; gas ratio; graphic images; dynamics of change; regression analysis; control levels; diagnostic distance
• Online Access: http://ee.zntu.edu.ua/article/view/123306

Purpose. Investigate the nature of the dynamics of changes in the criteria used for interpreting the results of DGA, in serviceable high-voltage transformers, leaky performance. Methodology. Theory of time series, regression analysis, the theory of pattern recognition, the method of reference levels, metric methods of recognition, diagnostics by distance to the standard. Findings. According to the results of the research, it is established that in normal functioning transformers, the values of all the diagnostic features used to interpret DGA results are changed randomly. Emergency actions on the part of the power grid lead to a short-term occurrence of a systematic component in the dependencies of the concentrations and rates of gas build-up on the duration of operation, and to a short-term stabilization of the values of the gas ratios at the level corresponding to this energy impact, as well as to the similarity of the graphic images. Originality. The performed analysis showed that in transformers of leaky performance, the appearance and development of a defect is accompanied not only by a change in the numerical values of the diagnostic criteria, which is known and used in the diagnosis, but also to a significant change in the nature of the dependencies of the diagnostic criteria versus time. Practical value. The obtained results make it possible to detect developing defects in transformers of non-germetic execution at an early stage of their development, even before the values of gas concentrations exceed the boundary values, which will help to avoid the destruction of insulation, and also to recognize the growth of concentrations of gases dissolved in oil, caused by the influence of emergency operation of electrical networks.