DGA analysis of transformer using Artificial neutral network to improve reliability in Power Transformers

Abstract

In electrical power system transformers are the most important elements. Any fault or damage in Transformer may interrupt continuous operation of electrical power system, as well as incur the expensive repair cost. Thus, it is necessary to conduct periodic inspections and maintenance for detection of incipient faults in power transformer to improve efficiency. Various off-line and on-line oil tests for fault diagnoses of power transformers are perform periodically as per expert recommendation. A number of standards have evolved over the time on transformer loading and power transformer fault diagnosis to minimize unplanned outages. Dissolve gas analysis is successful technique for identifying the incipient fault in a power transformer by analyzing ratios of dissolved gas concentrations arising from the deterioration of transformer liquid/solid insulations.In this paper multi layer perceptron type of artificial neural network is used with DGA methods to improve the reliability, efficiency and to increase power transformer life period. There is always problem in fault interpretation of multi Classification. ANN automatically tune the network Parameters, connection weights and bias terms of the neural networks to achieve the best model based on the proposed evolutionary algorithm, which provides the solution for complex classification problems DGA method find faults but during complexclassification it cannot give accurate results. To avoid such a conditions ANN is used with DGA in power transformer .