Article Preview
TopIntroduction
Due to the iterative development of new power systems and the rapid expansion of power grid scale, increasingly stringent requirements are being placed on the reliability and safety of electrical equipment, especially high-voltage switchgear. Gas insulated metal enclosed switchgear (GIS), as a core component of the power grid, has been widely adopted in power systems due to its advantages of compact size and high reliability (Liu et al., 2019; Wen et al., 2019). However, due to issues such as insulation material aging, power system overload, and insulation defects, various discharge phenomena, including partial discharge, corona discharge, surface discharge, and breakdown discharge, can occur within GIS. These discharges can easily lead to insulation breakdown, equipment failure and fire, and electromagnetic interference, posing significant threats to the stable operation of the power system (Ren et al., 2021; Sun et al., 2024). Different types of discharge phenomena generate signals of varying intensities. For instance, partial discharge typically produces low-intensity signals, while breakdown discharge generates high-intensity signals (Bassma & Tayeb, 2018; Behrmann & Smajic, 2016; Jangra et al., 2023; Ma et al., 2024; Song et al., 2022). Therefore, to ensure the safe and reliable operation of GIS equipment, it is imperative to explore high-accuracy discharge detection methods for discharge phenomena with varying signal intensities.
Traditional internal discharge detection methods for GIS based on acoustics and electronics are vulnerable to issues such as signal attenuation and high complexity of data collection and analysis, leading to lower accuracy and sensitivity of discharge detection (Kawakami et al., 2021; Li et al., 2018; Liu et al., 2014; Zhang et al., 2024). Bionic visual perception technology, by mimicking the function and principles of the biological visual system, can capture the optical signals radiated from GIS partial discharge (Zheng & Wu, 2021). Based on image processing and pattern recognition algorithms, it analyzes the characteristics and states of these signals, enabling real-time detection and identification of GIS internal partial discharge phenomena in complex electromagnetic environments (Yuwei et al., 2023). In the field of GIS internal discharge detection, bionic visual perception methods have been extensively studied due to their strong anti-interference capability, high sensitivity and accuracy, and adaptability to GIS internal environments. Li et al. (2023x) proposed a bionic visual perception method for GIS discharge signal detection with optical fiber rod, whereby a fault diagnosis model was established for GIS discharge optical signals by analyzing the temporal signal characteristics of partial discharge optical signals. This improved the effectiveness and diagnostic reliability of GIS partial discharge detection. Ren et al. (2021) proposed a bionic visual perception method for GIS by using an ultra-sensitive three-band optical local discharge sensor, which achieved efficient clustering of multiple discharges and accurate identification of discharge phenomena without the need for signal intensity and phase resolution statistics. Song et al. (2022) proposed a multi-scale fusion bionic visual perception method for simulating GIS discharge signals, whereby micro-scale optical simulation was combined with macro-scale circuit simulation models to improve the accuracy of GIS partial discharge signals perception under temperature effects. Lu et al. (2022) proposed a bionic visual sensing method for detecting GIS discharge signals, whereby sensors based on bionic visual perception were employed to enable comprehensive acquisition of information of discharge pulse signals and enhance the sensitivity of GIS discharge detection. However, due to factors such as reflections and obstructions inside GIS equipment, these detection methods face challenges including long detection times and susceptibility to environmental changes, leading to lower accuracy in imaging GIS internal discharge phenomena.