Abstract
Fault prediction for gas-insulated apparatus poses a critical challenge in the field of condition monitoring. This paper presents a GHz-bandwidth current and optoelectronic measurement system that is designed to monitor the broadband pulse currents and optical pulses of typical gas-insulated defect discharges as they approach critical breakdown. It statistically analyzes the frequency evolution of compact multi-peak pulse clusters at the nanosecond timescale during discharge progression, concluding that their proportion and statistical characteristics vary as the discharge develops. The paper investigates this phenomenon by providing a physical explanation for the generation mechanism of multi-peak pulse clusters and examining their potential as a criterion for insulation failure. In contrast to traditional phase-resolved pulse sequence analysis or ultra-high frequency diagnostics that rely on macroscopic pulse statistics or band-limited electromagnetic responses, the proposed GHz-bandwidth method directly captures nanosecond multi-peak pulse clusters associated with streamer branching, charge redistribution, and precursor-type leader initiation. This work presents a novel method for the early detection of insulation defects in gas-insulated equipment that is more closely aligned with the fundamental physics of discharges, with the potential to overcome the constraints of conventional phase-resolved statistics in the assessment of discharge hazards.
Funding Information
  • State Grid Corporation of China
  • Science and Technology Project (5500-202218131A-1-1-ZN)

This publication has 29 references indexed in Scilit: