悬链式单点系泊需要建立基于基础条件、工作条件、自存条件、运动和力要求等输入的模拟环境，并进行多点测试来寻找最佳设计。通过仿真计算构建2个主要数据集，即oper-ation数据集和self数据集。对self数据集进行预测，并将数据分为局部、全局和全局加局部3类进行训练和验证，使用4层全连接神经网络来预测回归问题，准确率可达90%以上。将该模型应用于更复杂的operation数据集时的效果并不理想。采用DNN+BN+ReLU作为最小分量自建模型DBRNet12复杂网络处理operation的数据得到86%的平均准确率。依据残差思想在DBRNet12基础上自建RNet40网络取得了90%的平均准确率。在网络架构方面，搭建了深度神经网络，通过全连接层进行预测，并对网络结构进行了持续的优化。最后，通过相对误差的评估来衡量预测效果的优劣，并利用残差网络进行优化。
A catenary single-point mooring requires a simulation environment based on inputs such as basic conditions, operating conditions, self-storage conditions, motion and force requirements, and multi-point testing to find the optimal design. The paper uses deep learning method to solve the mooring system prediction problem. Firstly, two datasets, i.e., self dataset and operation dataset are acquired by simulation calculation. Then the self dataset is predicted and the data is divided into three categories: local, global, and global plus local for training and validation, and a two-layer full-connected neural network is used to predict the regression problem with an accuracy of over 90%. As the results are not satisfactory when the model is applied to more complex operation datasets, a self-built model DBRNet12 complex network using DNN+BN+ReLU as the minimum component is added to handle more operation data, thus obtaining an average accuracy of 86%. The self-built RNet40 network based on the idea of residuals on DBRNet12 achieves a 90% average accuracy. In terms of network architecture, a deep neural network is built to predict parameters through fully connected layers, and the network structure is continuously optimized. Finally, the evaluation of relative error is used to evaluate the effectiveness of the prediction and the residual network is used for optimization. Through this procedure, the application effect of deep learning methods in mooring system prediction problems is achieved, and the ideas provide references for further research and practice in this field.
[1] 武文华,赵梓舒.基于深度学习的软刚臂系泊系统铰节点损伤识别方法研究[D].大连:大连理工大学, 2020.
[2] 邓林青,朱耀文,王宏伟,等.基于NARX神经网络的FPSO系泊缆张力预报[J].海洋工程装备与技术, 2020, 7(2):85-92.
[3] 张隆辉,刘正锋,魏纳新,等.基于NARX网络模型的系泊缆截断点运动计算[J].船舶力学, 2019, 23(8):936-947.
[4] 许小颖,周盼,王宽.基于神经网络和遗传算法的系泊优化设计[J].中国舰船研究, 2017, 12(5):97-103.
[5] 孙丽萍,张伟,艾尚茂.基于神经网络和遗传算法的系泊线长度参数优化[J].船舶工程, 2015, 37(4):88-91.
[6] 无深度学习技术下的人工智能时代[J].大数据时代, 2021(6):56-76.
[7] Shi Q, Chen W Y, Huang S Q, et al. Deep learning for mining protein data[J]. Briefings in Bioinformatics, 2021, 22(1):194-218.
[8] Elbrachter D, Perekrestenko D, Grohs P, et al. Deep neu-ral network approximation theory[J]. IEEE Transactions on Information Theory, 2021, 67(5):2581-2623.
[9] Lou R R, Lü Z H, Dang S P, et al. Application of ma-chine learning in ocean data[J]. Multimedia Systems, 2023, 29(3):1815-1824.
[10] Rohmer J, Idier D, Pedreros R. A nuanced quantile ran-dom forest approach for fast prediction of a stochastic marine flooding simulator applied to a macrotidal coastal site[J]. Stochastic Environmental Research and Risk As-sessment, 2020, 34(6):867-890.
[11] Melland P, Albright J, Urban N M. Differentiable pro-gramming for online training of a neural artificial viscosi-ty function within a staggered grid Lagrangian hydrody-namics scheme[J]. Machine Learning:Science and Tech-nology, 2021, 2(2):025015.
[12] Thiel C, Neumann K, Ludwar F, et al. Coating damage localization of naval vessels using artificial neural net-works[J]. Ocean Engineering, 2019, 192:106560.
[13] 周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报, 2017, 40(6):1229-1251.
[14] Zhang J, Xiao W D, Li Y J, et al. Residual compensa-tion extreme learning machine for regression[J]. Neuro-computing, 2018, 311(C):126-136.
[15] Garbin C, Zhu X Q, Marques O. Dropout vs. batch nor-malization:An empirical study of their impact to deep learning[J]. Multimedia Tools and Applications, 2020, 79(19):12777-12815.
[16] Bawa V S, Kumar V. Linearized sigmoidal activation:A novel activation function with tractable non-linear char-acteristics to boost representation capability[J]. Expert Systems with Applications, 2019, 120:346-356.
[17] Aydemir M E, Aydemir C. Residual displacement estima-tion of simple structuresconsidering soil structure inter-action[J]. Earthquakes and Structures, 2019, 16(1):69-82.
