engleski

Automatic Diagnosis of Power Transformers Based on Dissolved Gas Analysis – First Level of Diagnosis Using VAC and VEV Inference Methods

Assessment of power transformer condition became very important for utilities, to ensure continuous power transmission and power supply. Therefore, different techniques are in use for better condition assessment, as offline diagnostics and on-line monitoring systems. The off-line diagnostics has some time period between consecutive diagnoses, and during this period the equipment condition is unknown. Diagnostic tools in monitoring system usually comprise comparison of values of monitored quantities to preset limits, and alarming if these limits are exceeded. In this way weak diagnostic capabilities are achieved. Therefore, a new diagnosis model for assessment of condition of oil immersed power transformers was developed. This model is aimed to continuously and automatically diagnose transformer condition. Principle of the diagnostics is interpretation of dissolved gas analysis results (DGA) using several standardized interpretation methods. Then, on the basis of obtained diagnosis an overall diagnosis is inferred using VAC or VEV inference methods in a similar way as it is done by the human diagnostician. The diagnostic model shows excellent application flexibility, high robustness and acceptable diagnostic accuracy.

automatic diagnosis; power transformer; inference method; dissolved gas analysis

nije evidentirano