Partial discharges in gas insulation defects

RELEVANCE of the study is to obtain new experimental data to improve the reliability of electrical equipment diagnostics using partial discharge (PD) measurements. THE

Purpose

To determine the characteristics of partial discharges occurring within a single air defect located in a solid insulation.

Methods

To achieve this objective, high-voltage techniques were used to generate and experimentally measure high voltages that cause PDs in gas defects in construction. The field strength in gas defects in solid connections was calculated using the finite element method in COMSOL Multiphysics 6.0. Electrical circuit modeling in MultiSim, along with methods, capabilities, and statistics, were used to process the measurement results.

Results

This article presents a methodology for measuring the parameters of only those partial discharges (PDs) that occur within a single air-filled insulation defect using the contact electrical method. The electric field non-uniformity coefficient was calculated using the finite element method for samples with gas defects of varying sizes. Modeling in Multisim software was used to determine the phase shift angle between the current and voltage in the cell during PD measurements. For a measuring cell in which the electrodes are shaped like a spherical segment with a radius of 25 mm, the expected PD parameters were calculated depending on the insulation defect sizes. It was experimentally demonstrated that PDs in a sample with a single gas defect occur only twice per period of a sinusoidal industrial frequency voltage in the first and third quarters, with the electric field strength inside the defect having the same absolute value. No effect of the space charge generated in the defect cavity during each PD on the subsequent PD occurring halfway through the sine wave period was observed. No changes in the PD initiation voltage (Upd), pulse repetition rate, or PD amplitude were observed when the samples were held under a voltage of U~1.5*Upd for 400 hours. No signs of insulation degradation in the defect cavity walls were detected during this period.