Diagnostic evaluation of power transformers by an expert system

Power transformers are the most important and expensive equipment in the electrical transmission systems. The transformers average age in the power system in Hungary is approximately 25 years. It is not economical to replace these equipment only because of their age. We have to know and manage more information about the insulation, because the status of the insulation is determines the expected lifetime. A transformer can only be operated efficiently, reliably and economically, if we know its actual state expanding on all essential information. The operational reliability is influenced the most by the quality of the insulation but even other state markers/features play an important role. The quality of the insulation depends on the mechanical (temperature, vibration etc.) and electrical (voltage level, loading) influences. Besides these parameters there are a lot of things what affect the status of the insulation. These are the weather and the location of the equipment in the grid (rural or an industrial location). This paper is a review of an expert system which considers the parameters listed above. This expert system is developed in the High Voltage Laboratory of the Budapest Technical University and Economics. This evaluating system makes a technical and economical evaluation and comparison by marking the status of the transformers based on only a few parameters (status and the reliability of the status). This status contains the environmental influences, the operating values and the diagnostic results (Recovery Voltage Measurement (RVM), Dissolved Gas Analyzing (DGA) Oil test results etc.). The status can be calculated in the following way. Each physical effect (e.g.: oil test, RVM method etc.) has a value, and these physical effects are divided to physical values. These values are classified (e.g.: H2 content in the oil is 5 μl/l) to a classification value with their weight factor. Then the status code is calculated for the status attributes, with the classification value and a weighting. The global status factor is calculated with the status attributes (DGA, RVM, Oil, etc.) and their weight factor. The status diagram contains more status value and drawing these points in the function of the time, this diagram shows the changing of the status of the transformer. With the status diagram and with technical and statistic support, the failure level and some important information can be enumerated in the status system to manage the transformers lifecycle. In the electricity systems it is very important for the operating these equipments to have an expert system with suitable efficiency so the past record is also taken into consideration during the diagnoses.

[1]  R. Piercy,et al.  An Approach to Determine the Health Index of Power Transformers , 2008, Conference Record of the 2008 IEEE International Symposium on Electrical Insulation.

[2]  T. Hjartarson,et al.  Predicting Future Asset Condition Based on Current Health Index and Maintenance Level , 2006, ESMO 2006 - 2006 IEEE 11th International Conference on Transmission & Distribution Construction, Operation and Live-Line Maintenance.

[3]  Balint Nemeth,et al.  Condition monitoring of power transformers using DGA and Fuzzy logic , 2009, 2009 IEEE Electrical Insulation Conference.

[4]  Balint Nemeth,et al.  Transformer condition analyzing expert system using fuzzy neural system , 2010, 2010 IEEE International Symposium on Electrical Insulation.

[5]  L. L. Lai,et al.  A fuzzy dissolved gas analysis method for the diagnosis of multiple incipient faults in a transformer , 2000 .