3상 DC-AC 컨버터용 필름 커패시터 내부 상태 추정 기법 및 전해 커패시터와 비교 연구

Abstract

Metalized polymer-film capacitors have acquired a distinctive position among a variety of capacitor types due to their self-healing ability. Therefore, these devices are appropriate for critical power applications that demand reliability and durability. Nevertheless, as converters are increasingly being used for transmissions in networks, it is essential to improve stability to ensure the safety of system operations. Therefore, it is necessary to have a monitoring process that enables predictive maintenance to evaluate the health status and ensure the stability of electrical systems. However, the research in this field concentrates on electrolytic capacitors; and the characteristics of electrolytic and film capacitors differ. Hence, the need for further investigation into film capacitors is evident. This research proposes a condition monitoring approach that employs frequency signal analysis to assess the health status of capacitors in a three-phase AC-DC converter. The capacitor current is subjected to discrete wavelet transform and normalized by various indices, which serve as the input for learning algorithms. In addition, for comparison, capacitor voltage, output current, and output voltage are investigated using the discrete wavelet transform and fast Fourier transform. In this study, various indexes including root-mean-squared value, variance, average, and median, are utilized as inputs for artificial intelligent models to investigate factors affecting film capacitors. Eight learning algorithms are implemented to monitor the health status of film capacitors. The results show that utilizing the discrete wavelet transform combined with indexes for capacitor current yields a high accuracy of approximately 99.85%. These findings offer valuable insights into monitoring film capacitors using advanced techniques, and are anticipated to be informative for practical applications of film capacitor monitoring.