Prediction Model of Submarine Cable Burial Depth Trend Based on CNN-LSTM

In the process of submarine cable working, the fault caused by the change of buried depth becomes more and more prominent. However, because the change of submarine cable burial depth is difficult to detect, a prediction method of submarine cable burial depth change trend based on convolutional neural network and long short-term memory unit combined neural network is proposed. First established considering dynamic characteristics of submarine cable material thermal resistance, thermal road model, according to the cable core temperature data, calculate the submarine cable burial depth data sets, then using convolution neural network mining cable buried depth data set that is associated with the time node characteristic vector and the input results into long short-term memory in the network optimization training. Thus, the development trend of submarine cable buried depth can be predicted. Finally, the feasibility and effectiveness of the proposed method are verified by the prediction trend analysis of submarine cables buried in an offshore oil production platform. Compared with single convolutional neural network model and single long and short term memory neural network model, this method has higher prediction accuracy and higher prediction efficiency.