Improved Identification Method for Equivalent Network Parameters of Transformer Windings Based on Driving Point Admittance

Abstract

To accurately and efficiently determine the operating state of transformers, based on the driving point admittance method, a trapezoidal equivalent network model for dual-windings transformers was established. Based on Kirchhoff’s law, the node voltages and branch currents of the equivalent network model are obtained, and then the state space equations describing the network model parameters and driving point admittance data are obtained. The state space equation of the equivalent network model was constructed. Then, an improved whale optimization algorithm was proposed for the identification of parameters in the transformer equivalent network model. Random populations were generated using chaotic mapping, and the performance of the algorithm was improved by changing nonlinear control and adding adaptive weight coefficients. An objective function that simultaneously includes amplitude and phase frequency information at resonance points was established. Based on the improved whale optimization algorithm, the parameters of the equivalent network model were inverted. Finally, a comparative simulation test was conducted between the proposed method and existing main methods such as GA and PSO. The results indicate that the minimum objective value for parameter identification using the IWOA is merely 0.71, indicating that the IWOA possesses excellent parameter identification capabilities.