Abstract：This paper proposes a shallow buried state recognition method for submarine cables based on bitterling fish optimization (BFO)-variational mode decomposition (VMD)-extreme learning machine (ELM) fusion method to address the problems of low overall convergence speed, poor global search ability, low computational accuracy, and poor performance in handling nonlinear and non-stationary data in existing submarine cable depth recognition models. Firstly, the VMD method is introduced to decompose the raw data into multiple sub signals, which can effectively extract components of different frequencies and amplitudes, thus better capturing the characteristics of temperature data. Secondly, the underlying model is constructed by utilizing the fast training speed and good generalization ability of ELM algorithm. Finally, the BFO algorithm is used to optimize the VMD-ELM algorithm, improving convergence speed and global search capability of the model, and reducing the potential high computational complexity when processing large-scale data. The training set is input into the network for training, the effectiveness of the network is verified with the test set, and shallow buried state recognition of submarine cables is achieved. Through on-site collection of temperature data and burial depth measurement data of submarine cable optical fibers, the testing accuracy is high and the results show that this method can accurately identify the shallow burial state of submarine cables.
靳江江, 冯俊生, 李季聪, 唐佳. 基于温度特征分析的海缆埋深状态识别[J]. 电气技术, 2025, 26(8): 44-49.
JIN Jiangjiang, FENG Junsheng, LI Jicong, TANG Jia. State recognition of submarine cable burial depth based on temperature characteristic analysis. Electrical Engineering, 2025, 26(8): 44-49.
2025, Vol. 26 
