Research on broadband equivalent modeling for the dielectric response of transformer bushings based on the improved artificial bee colony algorithm

Abstract

As a new nondestructive diagnosis method of insulation state, frequency domain spectroscopy (FDS) has been widely used in the field detection of power equipment. However, the quantitative correlation between FDS measurement results and insulation state is not clear for transformer bushing, which seriously affects its detection effect. In order to solve this problem, the FDS modeling and parameter identification of transformer bushing has been studied in this paper. Firstly, the transformer bushing model was built in the laboratory, and the FDS measurement data of transformer bushings with different water content were collected. Then, a fusion algorithm of improved artificial bee colony algorithm and sequential quadratic programming algorithm was used to identify the parameters of transformer bushing model. After modeling and analyzing the FDS measured data, it is found that there is a good consistency between the measured spectrum of complex capacitance and the reconstructed spectrum of the model, and the goodness of fit of the model to the measured FDS exceeds 0.95. In addition, the identified values of resistance and capacitance in the equivalent model are within the reasonable range of 109Ω and 10−9F respectively. These results show that the extended Debye model and parameter identification algorithm can accurately model the frequency domain dielectric response of transformer bushing in a wide frequency range, and can provide an effective tool for more accurate analysis of transformer bushings insulation state.

Keywords

• Frequency domain spectroscopy
• Transformer bushings
• Insulation state
• Extended Debye​ model
• Parameter identification