Intermittent fault diagnosis in connector components based on arc wave characteristics

Abstract

Intermittent faults are widely present in aviation electronic devices, especially in various electrical connectors. It is usually hard to diagnose the source of the intermittent faults, which brings a huge challenge to the repair and maintenance of equipment. This paper focuses on the intermittent faults in typical aviation electrical connectors activated by shock test. The transient arc wave is observed on a nanosecond scale during the occurrence of the intermittent faults. An arc signal model is constructed to analyze the impact factors of the signal. Based on the arc wave characteristics, further intermittent fault diagnostic analyses are conducted on four types of connector components: damaged solder joints, cracked pin connections, loose wire connections and worn electrical connectors. The effective arc wave components of the raw signals are extracted using Variational Mode Decomposition (VMD), and a comparison is made between traditional diagnostic method and CNN-based deep learning method. The results show that the combination of VMD-CNN-SVM achieves the optimal diagnostic effect. The diagnostic results reflect that the proposed arc signal features are suitable for diagnosing intermittent faults in connector components.