Nonlinear photon emission and discharge mode transitions in HVDC needle–needle electrodes: Insights from UV photon spot distribution mapping

This study proposes a methodology for the development of ultraviolet (UV) spot distribution mapping with probabilistic processing, with the objective of addressing the challenges associated with the monitoring of partial discharge (PD) in high-voltage direct current (HVDC) equipment and analyzing its mechanism. An experimental platform with adjustable needle–needle electrodes (5–25 mm gap) was established to simulate insulation defects, enabling quantitative analysis through a dedicated UV image processing framework. The research demonstrates that the transition of photon counting characteristics from bell-shaped to U-shaped distributions with increasing electrode distance is governed by the spatial reorganization of discharge regions. UV spot distribution mapping reveals that this nonlinear behavior stems from expanded UV emission zones at larger distances and constrained discharge region coalescence at smaller gaps, while spectral red-shift toward infrared wavelengths during discharge intensification further modulates photon counting patterns. A unified mathematical model was developed, incorporating both spatial and spectral evolution mechanisms. The unified model, designated as, has been shown to be effective in this regard. Spatial analysis identified characteristic discharge core zones with pronounced polarity asymmetry. The probabilistic methodology effectively discriminates between persistent discharge zones and random photon events, enabling quantitative assessment of discharge stability and evolution. The findings establish a framework correlating UV spot distribution dynamics with photon counting behavior, providing both fundamental insights into DC PD mechanisms and practical methodologies for insulation condition assessment in HVDC systems.