Intelligent protection scheme for power transformer

This paper presents the development of a small-scale model of a single-phase transformer differential protection scheme in a laboratory environment. The developed model is realistic in nature as it closely resembles various phenomena that occur in an actual transformer. Moreover, the above model is beneficial to undergraduate/post-graduate students, faculty members and utility engineers as they would get a real feel of the actual working of the transformer. Initially, mathematical validation of the hardware setup is carried out for various operating conditions of the transformer. Based on these mathematical calculations, the settings of the digital relay have been done. Thereafter, the performance of the proposed setup has been evaluated on various operating conditions of the transformer such as no-load/full load condition, internal faults, external faults and magnetizing inrush. In addition, the paper proposes new Probabilistic Neural Network (PNN) based adaptive scheme to rectify unintended operation of the conventional digital power transformer protection relays. PNN is able to provide effective discrimination between internal fault and external disturbances with an accuracy of 99.42%.