Ship track prediction based on Bayesian optimization in temporal convolutional networks

Table 1. MAE evaluation indicators for pre-processed data
View table
View in Article
Table 1. MAE evaluation indicators for pre-processed data
航迹类型预处理前预处理后
经度/（°）纬度/（°）经度/（°）纬度/(°)
直线型航迹 0.0012 0.0011 0.0008 0.0007
S型航迹 0.0058 0.0063 0.0051 0.0056
M型航迹 0.0066 0.0069 0.0057 0.0062

Table 2. Parameters for experimental model
View table
View in Article
Table 2. Parameters for experimental model
算法参数 TCN LSTM
卷积核大小 5
卷积核数量 64
扩张因子 8
优化器 Adam Adam
隐藏层 8 64
学习率 0.001 0.001
迭代次数 1 000 1 000
批次大小 64 64

Table 3. Results comparison of crossover experiments

View table

View in Article

Table 3. Results comparison of crossover experiments

算法模型	${e_{\rm{RMSE}}}$		${e_{\rm{MAE}}}$
算法模型	经度/（°）	纬度/（°）	经度/（°）	纬度/（°）
直线型航迹	0.0008	0.0003	0.0002	0.0006
S型航迹	0.0024	0.0021	0.0023	0.0025
M型航迹	0.0024	0.0035	0.0018	0.0033

Tools

Get Citation

Copy Citation Text

Jinyuan LI, Faxin ZHU, Xianbin TENG, Qilin BI. Ship track prediction based on Bayesian optimization in temporal convolutional networks[J]. Chinese Journal of Ship Research, 2024, 19(6): 303

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites