基于ASAE深度学习预测海洋气象对船舶航速的影响

王胜正; 申心泉; 赵建森; 冀宝仙; 杨平安

doi:10.19818/j.cnki.1671-1637.2018.02.015

基于ASAE深度学习预测海洋气象对船舶航速的影响

doi: 10.19818/j.cnki.1671-1637.2018.02.015

上海海事大学机构商船学院, 上海 201306

基金项目:

国家自然科学基金项目 51709167

国家自然科学基金项目 61304230

上海市曙光人才计划 15SG44

上海海事大学研究生学术新人培育项目 YXR2016089

详细信息

作者简介:
王胜正(1976-), 男, 湖南双峰人, 上海海事大学教授, 工学博士, 从事船舶智能航行研究

中图分类号: U675.79
计量
- 文章访问数: 725
- HTML全文浏览量: 229
- PDF下载量: 1006
- 被引次数: 0
出版历程
- 收稿日期: 2017-11-02
- 刊出日期: 2018-04-25

Prediction of marine meteorological effect on ship speed based on ASAE deep learning

Merchant Marine College, Shanghai Maritime University, Shanghai 201306, China

More Information

Author Bio:
WANG Sheng-zheng(1976-), male, professor, PhD, szwang@shmtu.edu.cn

摘要

摘要: 为了有效地预测海洋气象对船舶航速的影响, 在稀疏自编码(SAE) 网络模型的基础上提出交替稀疏自编码(ASAE) 网络模型; 构建了海洋气象对船舶航速影响的预测框架, 利用关联规则方法对航行数据进行特征选择, 挖掘了船速影响因素及其隐含关系; 整合了中国远洋海运集团有限公司提供的船舶航行数据以及美国国家海洋和大气管理局提供的气象数据, 用训练样本对ASAE网络模型进行训练, 用测试样本对ASAE网络模型进行验证, 并与支持向量回归(SVR) 模型、反向传播神经网络(BPNN) 模型、深度信念网络(DBN) 模型及SAE网络模型的预测结果进行了对比。研究结果表明: ASAE网络模型的训练时间和海洋气象对船舶航速影响预测值的均方根误差分别为8.2s和0.287 3kn, 与SVR模型、BPNN模型、DBN模型及SAE网络模型相比, 训练时间分别缩短了1 683.1、66.9、2.0、1.5s, 预测准确度分别提高了0.045 5、0.296 9、0.153 4、0.178 6kn; ASAE网络模型的预测结果更符合实际海况, 可动态掌握海洋气象对船舶航速的影响; 通过预测的航速影响值来推算实际航速可为气象导航优化船舶运输过程起到辅助作用, 在进行航线规划、航速推荐等航行优化策略时能准确考虑海洋气象所产生的复杂影响, 从而改善船舶运营能效指标, 实现节能、低碳、绿色航行的宗旨。
- 交通信息工程 /
- 智能航行 /
- 船舶航速 /
- 深度学习 /
- 交替稀疏自编码 /
- 关联规则 /
- 气象因子
Abstract: In order to effectively predict the marine meteorological effect on ship speed, the alternating sparse auto-encoders (ASAE) network model based on the sparse auto-encoders (SAE) network model was proposed.A framework for predicting the marine meteorological effect on ship speed was constructed, and the association rules method was proposed to feature the navigation data, so as to excavate the factors and implicit relations affecting ship speed.Through the integration of ship navigation data provided by China COSCO Shipping Group and the meteorological data provided by the National Oceanic and Atmospheric Administration, ASAE network model was trained with training samples and verified with test samples, and theprediction result was compared to the results gained by support vector regression (SVR) model, back propagation neural network (BPNN) model, deep belief network (DBN) model and SAE network model.Research result shows that the training time and the mean squared error of marine meteorological effect on ship speed gained by ASAE network model are 8.2 sand 0.287 3 kn, respectively.Compared to SVR model, BPNN model, DBN model, and SAE network model, ASAE network model can shorten the training time by 1 683.1, 66.9, 2.0 and 1.5 s, respectively, and can increases the prediction accuracy by 0.045 5, 0.296 9, 0.153 4 and 0.178 6 kn.The forecast result of ASAE network model is more in line with actual sea condition, and can dynamically master the marine meteorological effect on ship speed.Estimating the actual speed through the predicted values can optimize the ship transportation process in meteorological navigation.It plays an auxiliary role to accurately consider the complex impacts of marine meteorological on navigation optimization strategies such as route planning and speed recommendation.Thereby, it improves the energy efficiency indicator of ship operation and achieves the purpose of energy saving, low-carbon and green navigation.
- traffic information engineering /
- intelligence voyage /
- ship speed /
- deep learning /
- alternating sparse auto-encoder /
- association rule /
- meteorological factor

HTML全文

图 1 SAE网络模型训练过程

Figure 1. Training process of SAE network model

下载: 全尺寸图片幻灯片

图 2 ASAE网络模型训练过程

Figure 2. Training process of ASAE network model

下载: 全尺寸图片幻灯片

图 3 海洋气象对船舶航速影响预测框架

Figure 3. Prediction framework of marine meteorological effect on ship speed

下载: 全尺寸图片幻灯片

图 4 规则强度指标

Figure 4. Rule strength indicators

下载: 全尺寸图片幻灯片

图 5 预测结果的累计百分数

Figure 5. Accumulated percentages of predicted results

下载: 全尺寸图片幻灯片

图 6 SAE和ASAE损失函数

Figure 6. Loss functions of SAE and ASAE

下载: 全尺寸图片幻灯片

表 1 货船COSCO ASIA的部分航行数据

Table 1. Part of voyage data of freighter COSCO ASIA

下载: 导出CSV

表 2 NOAA部分气象数据

Table 2. Part of metorological data of NOAA

下载: 导出CSV

表 3 部分原始训练数据

Table 3. Part of original training data

下载: 导出CSV

表 4 不同吃水的归一化值和特征向量

Table 4. Eigenvectors and normalized values of different drafts

下载: 导出CSV

表 5 各模型预测性能

Table 5. Prediction performances of each model

下载: 导出CSV

参考文献(23)

[1]	孙健, 王凤武, 刘强, 等. 基于证据理论的船舶大风浪中航行的安全评价[J]. 大连海事大学报, 2013, 39 (1): 53-56. https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201301013.htm SUN Jian, WANG Feng-wu, LIU Qiang, et al. Safety assessment of ships navigating in heavy sea based on evidence theory[J]. Journal of Dalian Maritime University, 2013, 39 (1): 53-56. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201301013.htm
[2]	俞姗姗, 汪传旭. 不同碳排放调控政策下的船舶航速优化[J]. 大连海事大学学报, 2015, 41 (3): 45-50. https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201503009.htm YU Shan-shan, WANG Chuan-xu. Speed optimization under different carbon emission control policy[J]. Journal of Dalian Maritime University, 2015, 41 (3): 45-50. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201503009.htm
[3]	孟晓东, 袁章新. 考虑不规则风浪影响的最小油耗航速模型[J]. 上海海事大学学报, 2016, 37 (1): 19-24. https://www.cnki.com.cn/Article/CJFDTOTAL-SHHY201601004.htm MENG Xiao-dong, YUAN Zhang-xin. A minimum fuel consumption speed model considering effect of irregular wind and wave[J]. Journal of Shanghai Maritime University, 2016, 37 (1): 19-24. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-SHHY201601004.htm
[4]	SHENG X M, CHEW E P, LEE L H. (s, S) policy model for liner shipping refueling and sailing speed optimization problem[J]. Transportation Research Part E: Logistics and Transportation Review, 2015, 76: 76-92. doi: 10.1016/j.tre.2014.12.001
[5]	KOBAYASHI E, HASHIMOTO H, TANIGUCHI Y, et al. Advanced optimized weather routing for an ocean-going vessel[C]//IEEE. Proceedings of 2015 International Association of Institutes of Navigation World Congress. New York: IEEE, 2015: 1-8.
[6]	LI Yuan-kui, ZHANG Ying-jun, GAO Zong-jiang, et al. Optimal routing model for wind-assisted ship[C]//IEEE. Proceedings of 2013International Conference on Mechatronic Sciences, Electric Engineering and Computer. New York: IEEE, 2013: 1567-1571.
[7]	LU R H, TURAN O, BOULOUGOURIS E, et al. A semiempirical ship operational performance prediction model for voyage optimization towards energy efficient shipping[J]. Ocean Engineering, 2015, 110: 18-28. doi: 10.1016/j.oceaneng.2015.07.042
[8]	VENETI A, KONSTANTOPOULOS C, PANTZIOU G. An evolutionary approach to multi-objective ship weather routing[C]//IEEE. Proceedings of the 6th International Conference on Information, Intelligence, Systems and Applications. New York: IEEE, 2015: 1-6.
[9]	SHI Bu-hai, GUO Xie-tao, ZHANG Ben. Research for economy shipping of oceangoing vessel based on the FOA-SVR[C]//IEEE. Proceedings of the 32nd Chinese Control Conference. New York: IEEE, 2013: 7563-7568.
[10]	LI Gang, XU Huan, LIU Wei. Option of operating speed for vessels under low-carbon economy[J]. Journal of Industrial Engineering and Management, 2013, 6 (1): 289-296.
[11]	LI Xiao-ming, XIAO Jian-mei, WANG Xi-huai. Optimization of ship routing with tabu search algorithm[C]//IEEE. Proceedings of 2011International Conference on Energy and Environment. New York: IEEE, 2011: 139-142.
[12]	CHEN Yuan-chao, WU Qi-rui, WANG Yu-cheng. A research on precise prediction method of high-speed craft resistance based on CFD[C]//IEEE. Proceedings of the 8th International Conference on Intelligent Computation Technology and Automation. New York: IEEE, 2015: 84-88.
[13]	YANG H F, DILLON T S, CHEN Y P P. Optimized structure of the traffic flow forecasting model with a deep learning approach[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, 28 (10): 1-11.
[14]	GAN Shao-jun, LIANG Shan, LI Kang, et al. Long-term ship speed prediction for intelligent traffic signaling[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 18 (1): 1-10.
[15]	SALMAN A G, KANIGORO B, HERYADI Y. Weather forecasting using deep learning techniques[C]//IEEE. Proceedings of 2015 International Conference on Advance Computer Science and Information Systems. New York: IEEE, 2015: 281-285.
[16]	KOESDWIADY A, SOUA R, KARRAY F. Improving traffic flow prediction with weather information in connected cars: a deep learning approach[J]. IEEE Transaction on Vehicular Technology, 2016, 65 (12): 1-10.
[17]	HOSSAIN M, REKABDAR B, LOUIS S J, et al. Forecasting the weather of Nevada: a deep learning approach[C]//IEEE. Proceedings of 2015International Joint Conference on Neural Networks. New York: IEEE, 2015: 1-6.
[18]	BENGIO Y. Learning deep architectures for AI[J]. Foundations and Trends in Machine Learning, 2009, 2 (1): 1-12.
[19]	JIANG Xiao-juan, ZHANG Yin-hua, ZHANG Wen-shen, et al. A novel sparse auto-encoder for deep unsupervised learning[C]//IEEE. Proceedings of 6th International Conference on Advanced Computational Intelligence. New York: IEEE, 2013: 256-261.
[20]	BENGIO Y, LAMBLIN P, POPOVICI D, et al. Greedy layer-wise training of deep networks[J]. Advances in Neural Information Processing Systems, 2007, 19 (3): 153-160.
[21]	FRIGUI H, NASRAOUI O. Unsupervised learning of prototypes and attribute weights[J]. Pattern Recognition, 2004, 37 (3): 567-581.
[22]	KARANCE L, ERDEM A, ERDEM E. Image matting with KL-divergence based sparse sampling[C]//IEEE. Proceedings of2015IEEE International Conference on Computer Vision. New York: IEEE, 2015: 7-13.
[23]	CAI C H, FU A W C, CHENG C H, et al. Mining association rules with weighted items[C]//IEEE. International Symposium on Database Engineering and Applications. New York: IEEE, 1998: 1-10.