Genetic algorithm–based health index determination of distribution transformer

Summary This paper presents a smart online condition monitoring system for distribution transformer monitoring and condition analysis. The system comprises of intelligent electronic device capable of receiving input from transformer through sensors. The processing unit calculates the health index ( HI ) of transformer based on weightage associated with healthy sensor input optimized by genetic algorithm. Use of communication system enables to initiate action based on command received from the substation. Thus, active control of the system is possible. In extreme cases like overload, overvoltage, excessive oil temperature, and low oil level, the algorithm can issue a local trip command. The experimentation is performed on 15‐kVA, 400/400 volts custom made transformer for different operating conditions. Health indices are computed by conventional numerical and genetic algorithm methods, which offer improved weighing approach than the former. The proposed system offers development of smart transformer for applications in the growing distribution grid.