Condition monitoring and fault diagnosis of single phase transformers (MS)

Abstract

Transformers are the most critical and capital intensive assets of electrical transmission and distribution system and their reliable operation is necessary for maintaining healthy functioning of power grid. Transformer failures could cause power outages, personal and environmental hazards and expensive rerouting or purchase of power from other suppliers. Transformer failures can occur due to various causes. Transformer in-service interruptions and failures usually result from dielectric breakdown, winding distortion caused by short-circuit withstand, winding and magnetic circuit hot spot, electrical disturbances, deterioration of insulation, lightning, inadequate maintenance, loose connections, overloading, failure of accessories such as OLTCs, bushings, etc. The condition of transformer windings can be gauged by monitoring their equivalent circuit parameters. These parameters are not affected by external faults and change only in the presence of an internal aberration. Changes in the insulation temperature are reflected in winding temperature and can be monitored by observing the winding resistance values. Similarly changes in the short circuit reactance can give information on the condition and structure of windings. Rapid and reliable protection can be implemented by monitoring these parameters since inrush current and over-excitation does not affect these parameters. Presently, there is no accurate measurement method for the transformer winding parameters and generally require the transformer to be disconnected from the power system. A new algorithm for extracting transformer winding parameters which can be implemented online is presented. This method takes only the input currents and voltages as inputs and thereby eliminates the need for the disconnection of the transformer from the power system. In this method, winding parameters are obtained by solving the equivalent circuit equations in real time continuously which allows for interpretation of transformer condition and detection of faults in real-time. The proposed method has been tested and validated by simulations and experiments. In thesis, a novel method is presented for the extraction of winding parameters of a single-phase transformer. The primary and secondary currents and voltages are taken as inputs, which are used to solve the discretized equivalent parameter equations online to find the winding parameters which allow monitoring of the condition while the transformer is in operation. Moreover it requires minimal computations. The proposed algorithm has been tested for various cases is found to be sufficiently accurate to monitor significant changes in the equivalent parameter values. Further the interpretation of variation observed in the winding parameters is explained with reference to the possible fault occurrence. It is shown how the parameters will vary when an inter-turn fault, insulation quality compromise and mechanical integrity compromise possibly occur. It was shown that inter-turn faults can be identified by decrease in resistive loss component, insulation quality can be monitored via winding resistance monitoring and mechanical integrity monitored by observing the variation in leakage flux losses.