摘要
海底观测网长期受海洋环境与人为因素影响,易使光电复合缆绝缘破损与海水接触形成电学故障点。如何准确地定位电学故障点,对提高海底观测网输电与信息传输的可靠性至关重要。首先根据海底观测网输电结构建立海底观测网输电模型,推导与模拟电学故障点传播至观测点的暂态电流,然后由连续小波变换提取暂态电流与故障点对应的内在关联特征量,最后通过Swin-Transformer神经网络挖掘内在关联特征量与故障距离的匹配关系来定位电学故障点。研究结果表明,在内在关联特征量样本测试集条件下,光电复合缆≤160 km的电学故障点定位误差小于400 m,可为长距离光电复合缆的海底观测网电学故障点定位提供参考。
The submarine observation network is often affected by the marine environment and human factors,leading to insulation damage in the photoelectric composite cable and the formation of electrical fault points in contact with seawater.Accurately locating electrical fault points is crucial for improving the reliability of power and information transmission in the underwater observation network.Firstly,a transmission model is established according to the transmission structure of the submarine observation network,and the transient currents propagating from the electrical fault points to the observation points are derived.Subsequently,continuous wavelet transformation is applied to extract intrinsic correlational features between transient currents and fault points.Finally,a neural network,Swin-Transformer,is utilized to explore the matching relationship between intrinsic correlational features and fault distances,thereby locating the electrical fault points.The research results indicate that under the conditions of the sample test set of intrinsic correlational features,the positioning error of electrical fault points in photoelectric composite cables in the length of 160 km is less than 400 m.This method provides an effective approach for locating electrical fault points in the submarine observation network of long-distance photoelectric composite cables.
作者
栾韶泽
李光炬
甘维明
季桂花
邢炜光
赵赞善
LUAN Shaoze;LI Guangju;GAN Weiming;JI Guihua;XING Weiguang;ZHAO Zanshan(Hainan Acoustics Laboratory,Institute of Acoustics,Chinese Academy of Sciences,Haikou,570105,China;University of Chinese Academy of Sciences,Beijing,100190,China;Lingshui,Marine Information,Hainan Observation and Research Station,Lingshui,572423,China;Key Laboratory of Ocean Observation Technology,Ministry of Natural Resources,Tianjin,300112,China)
出处
《网络新媒体技术》
2024年第3期47-56,共10页
Network New Media Technology
基金
海南省自然科学基金项目(编号:523QN309)
自然资源部海洋观测技术重点实验室开放基金项目(编号:2023klootA07)。
