In this paper, Object recognition of Partial Discharge (PD) occurring in High Voltage (HV) insulation system of 6.6kV, 1.5 MVA stator coils for different fault and fault conditions are analyzed using a technique of object defining using the matrix representation form and analyzing using the matrix represented Neural Networks (NN) i.e., Cellular Neural Networks (CNN) realized on a Digital Signal Processor (DSP) and data collected using PD Detector Biddle series 27000 along with Lemke’s software in numerical and pictorial form is used to compare the condition of the insulation system. Comparisons of the trained CNN output in numerical and pictorial comparison are presented in results with tabulated numerical results and error graphs.
Object Recognition (OR), Partial Discharge(PD), Cellular Neural Networks(CNN), Digital Signal Processor (DSP), High Voltage (HV), Normal Coil (NC), End-Winding discharge coil (EWDC), Slot discharge coil (SDC) & Semi-Conductor paint (SCP)