Theoretical and experimental investigation of transformer winding fault detection using comparison of transfer function coefficients

Abstract

In this work, a new model of transformer winding is developed. The components in the model are determined by the geometric and electric data of the winding (detailed model) and using experimental data based on genetic algorithm. Under different degrees of axial displacement and radial deformation in the winding, the circuit parameters of the model will change and thus the equivalent circuit characteristics will be influenced. After acquiring the model parameters in the intact and faulted cases, transfer function coefficients are derived in model using nodal analysis. Subsequently, introducing a new index based on these coefficients, the type and extent of penetration of the fault in the winding can be specified. Results presented in this work demonstrate the potential of the proposed method.