AUTHOR=Heng Qian , Ni Qi , Wang Biao , Ding Yi , Sun Meng , Ding Yang 

TITLE=High-temperature partial discharge characteristics of power module packaging insulation under square pulses based on multispectral optical detection method

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2025.1580823

DOI=10.3389/fenrg.2025.1580823

ISSN=2296-598X

ABSTRACT=The power module package insulation is prone to partial discharge (PD) under fast square pulses, which is one of the main reasons for power module packaging failure. However, there are few studies on PD under square electrothermal coupling stress. One of the important reasons is that the electromagnetic interference and displacement current generated by the rising and falling edges of the square pulse seriously interfere with the detection of PD. To address this problem, this paper proposes a multi-spectral optical detection method with high sensitivity and strong anti-interference. Based on the multi-spectral synchronous detection platform, the high-temperature PD characteristics of power module package insulation under square pulses are studied. The results show that the partial discharge inception voltage (PDIV) decreases with increasing temperature, and the intensity of the PD multi-spectral signal increases with increasing temperature. In addition, the statistical analysis of the maximum optical signal intensity, cumulative optical signal intensity, number of PDs, and phase of PD under square pulse shows that with the increase of temperature, the maximum optical signal intensity and cumulative optical signal intensity of PD increase, and PD events mostly occur on both sides of the square pulse. Among them, the optical signal in the near-ultraviolet band is dominant, and the near-infrared signal has an obvious increase trend. These findings are of great significance for the reliability detection of power module package insulation.