Transformer Parameters Estimation From Nameplate Data Using Evolutionary Programming Techniques

This paper proposes a simple and effective evolutionary computation-based technique to estimate the equivalent circuit parameters of a single-phase transformer from its nameplate data without the need to conduct any experimental measurements. Two techniques, namely: particle swarm optimization and genetic algorithm are employed to track nameplate data by minimizing certain objective functions. The effectiveness of the proposed technique is examined through its application for three single-phase transformers of different ratings. The results show that evolutionary computation techniques can precisely identify transformer equivalent circuit parameters. The proposed technique can be extended to estimate the parameters of a three-phase power transformer from its nameplate data without taking the transformer out of service to carry out any experimental testing.