基于LightGBM的船舶航速预测模型

朱晓晨; 尹奇志; 赵福芹; 钱巍文; 赵奎奎

doi:10.16411/j.cnki.issn1006-7736.2023.01.006

大连海事大学学报 >

2023 , Vol. 49 >Issue 1: 56 - 65

DOI: https://doi.org/10.16411/j.cnki.issn1006-7736.2023.01.006

交通运输工程

基于LightGBM的船舶航速预测模型

朱晓晨 ,
尹奇志 ,
赵福芹 ,
钱巍文 ,
赵奎奎

展开

(1.武汉理工大学 a.交通与物流工程学院/可靠性工程研究所；b.船海与能源动力工程学院；c.交通与物流工程学院，武汉430063；2. 潍柴动力股份有限公司，山东潍坊 261061)

朱晓晨（1997 — ），男，硕士生;尹奇志（1976 — ），男，博士，副教授，博士生导师， E-mail：qzyin@whut.edu.cn

收稿日期: 2022-10-14

修回日期: 2022-12-13

网络出版日期: 2022-12-13

基金资助

绿色智能内河船舶创新专项（装函2019）；潍柴动力股份有限公司技术项目（WCDL-GH-2021-0050）

收起

Ship speed prediction model based on LightGBM

ZHU Xiao-chen ,
YIN Qi-zhi ,
ZHAO Fu-qin ,
QIAN Wei-wen ,
ZHAO Kui-kui

Expand

(1a.Reliability Engineering Institute of School of Transportation and Logistics Engineering; 1b.School of Naval Architecture, Ocean and Energy Power Engineering; 1c.School of Transportation and Logistics Engineering, Wuhan University of Technology, Wuhan 430063, China; 2.Weichai Power Company Limited, Weifang 261061, China)

Received date: 2022-10-14

Revised date: 2022-12-13

Online published: 2022-12-13

Fold

摘要

针对现有基于机器学习算法的船舶航速预测模型无法兼顾计算精度高、泛化能力强及计算速度快的问题，提出基于LightGBM的船舶航速预测模型，并以一艘安装有能效监测系统的内河船舶为研究对象，运用LightGBM算法建立以实时风速、风向、水深、水流速度、尾轴转速、轴功率和主机油耗为输入的船舶航速预测模型，同时与RR、SVR、DT、BPNN、RF、GBDT和XGBoost七种机器学习算法的航速预测结果进行比较。结果表明：基于LightGBM建立的船舶航速预测模型的精度、泛化能力、运算速度均排名第二，综合性能最好，能在保证较高预测精度和较强泛化能力前提下，实现对船舶航速的快速预测。

关键词： 船舶; 航速预测; 机器学习; 回归分析; LightGBM

本文引用格式

朱晓晨 , 尹奇志 , 赵福芹 , 钱巍文 , 赵奎奎 . 基于LightGBM的船舶航速预测模型[J]. 大连海事大学学报, 2023 , 49(1) : 56 -65 . DOI: 10.16411/j.cnki.issn1006-7736.2023.01.006

Abstract

Aiming at the problem that the existing ship speed prediction model based on machine learning algorithm was impossible to balance the high calculation accuracy, strong generalization ability and fast calculation speed, a ship speed prediction model based on LightGBM (Light Gradient Boosting Machine) was proposed. Taking an inland river ship equipped with an energy efficiency monitoring system as the research object, the ship speed prediction model with real-time wind speed, wind direction, water depth, water speed, tail shaft speed, shaft power and main engine fuel consumption as inputs was established by using LightGBM algorithm, and compared with the speed prediction results of seven machine learning algorithms as RR (Ridge Region), SVR (Support Vector Region), DT (Decision Tree), BPNN (Back Propagation Neural Network), RF (Random Forest), GBDT (Gradient Boosting Decision Tree) and XGBoost (Extreme Gradient Boosting) at the same time. The results show that the ship speed prediction model based on LightGBM ranks second in accuracy, generalization ability and calculation speed with best comprehensive performance, which can realize fast prediction of ship speed on the premise of ensuring high prediction accuracy and strong generalization ability.

Key words： ship; speed prediction; machine learning; regression analysis; LightGBM

参考文献

[1]袁裕鹏,王康豫,尹奇志,等.船舶航速优化综述[J].交通运输工程学报,2020,20(6):18-34．

YUAN Y P, WANG K Y, YIN Q Z, et al. Review on ship speed optimization[J]. Journal of Traffic and Transportation Engineering, 2020, 20(6): 18-34. (in Chinese)

[2]范爱龙,王拯,孙星,等.基于不同场景的船舶航速优化模型与影响因素研究[J].中国造船,2021,62(1):162-171.

FAN A L, WANG Z, SUN X, et al. Study of ship speed optimisation model and influencing factors based on different scenarios[J]. Shipbuilding of China, 2021, 62(1): 162-171. (in Chinese)

[3]黄连忠,万晓跃,孙永刚,等.基于模拟退火算法的船舶航速优化研究[J].船舶,2018,29(增1):8-17.

HUANG L Z, WAN X Y, SUN Y G, et al. Research on ship speed optimization based on simulated annealing algorithm[J]. Ship & Boat, 2018, 29(Sup.1): 8-17. (in Chinese)

[4]MAO W G, RYCHLIK I, WALLIN J, et al. Statistical models for the speed prediction of a container ship[J]. Ocean Engineering, 2016, 126: 152-162.

[5]王志芳,阳子轩,陈庆任.数字化船舶EEDI验证方法[J].中国机械工程,2022,33(10):1189-1194.

WANG Z F, YAGN Z X, CHEN Q R. Digital vertification method of ship EEDI[J]. China Mechanical Engineering, 2022, 33(10): 1189-1194. (in Chinese)

[6]HU Z H, ZHOU T R, OSMAN M T, et al. A novel hybrid fuel consumption prediction model for ocean-going container ships based on sensor data[J]. Journal of Marine Science and Engineering, 2021, 9(4): 449.

[7]YOO B, KIM J. Powering performance analysis of full-scale ships under environmental disturbances[J]. IFAC-Papers OnLine, 2017, 50(1): 2323-2328.

[8]BASSAM A M, PHILLIPS A B, TURNOCK S R, et al. Ship speed prediction based on machine learning for efficient shipping operation[J]. Ocean Engineering, 2022, 245: 110449.

[9]SONER O, AKYUZ E, CELIK M. Statistical modelling of ship operational performance monitoring problem[J]. Journal of Marine Science and Technology, 2019, 24(2): 543-552.

[10]XU W G, LI Z W, HU Q, et al. Research on ship speed prediction model based on BP neural network[C]//International Conference on Bio-Inspired Computing: Theories and Applications. Singapore: Springer, 2021: 405-411.

[11]TARELKO W, RUDZKI K. Applying artificial neural networks for modelling ship speed and fuel consumption[J]. Neural Computing and Applications, 2020, 32: 17379-17395.

[12]ABEBE M, SHIN Y, NOH Y, et al. Machine learning approaches for ship speed prediction towards energy efficient shipping[J]. Applied Sciences, 2020, 10(7): 2325.

[13]张江帆.基于高斯过程回归模型的锂电池数据处理[D].北京:北京交通大学,2017.

ZHANG J F. Lithium-ion battery data processing based on Gaussian process regression model[D]. Beijing: Beijing Jiaotong University, 2017.(in Chinese)

[14]陈翌,白云飞,何瑛.数据驱动的锂电池健康状态估算方法比较[J].储能科学与技术,2019,8(6):1204-1210.

CHEN Y, BAI Y F, HE Y. Comparison of data-driven lithium battery state of health estimation methods[J]. Energy Storage Science and Technology, 2019, 8(6): 1204-1210. (in Chinese)

[15]张隅希,段宗涛,朱依水,等.机动车油耗模型研究综述[J].计算机工程与应用,2021,57(24):14-26.

ZHANG Y X, DUAN Z T, ZHU Y S, et al. Survey of fuel consumption model for motor vehicle[J]. Computer Engineering and Applications, 2021, 57(24): 14-26. (in Chinese)

[16]齐琦,陈芳芳,赵辉,等.基于优化BP神经网络光伏出力短期预测研究[J].计算机测量与控制,2021,29(4):70-75.

QI Q, CHEN F F, ZHAO H, et al. Prediction of photovoltaic output based on improved neural network[J]. Computer Measurement & Control, 2021, 29(4): 70-75. (in Chinese)

[17]唐若男. 基于机器学习的复杂场景杂波识别算法[D].成都:四川大学,2021.

TANG R N. A clutter recognition algorithm for complex scenes based on machine learning[D].Chengdu : Sichuan University, 2021. (in Chinese)

[18]熊壮.基于视觉感知特性与图像特征的图像质量评价[D].海口:海南大学,2020.

XIONG Z. Image quality evaluation based on visual perception and image features[D]. Haikou: Hainan University, 2020. (in Chinese)

[19]周治国,李怡瑶,屈崇,等.基于LightGBM的船用柴油机监测数据预测分析[C]//第十四届全国信号和智能信息处理与应用学术会议论文集.北京:《计算机工程与应用》编辑部,2021.

ZHOU Z G, LI Y Y, QU C, et al. Predictive analysis of monitoring data of marine diesel engine based on LightGBM[C]//Proceedings of the 14th National Conference on Signal and Intelligent Information Processing and Application. Beijing: Editorial Board of Computer Engineering and Applications, 2021. (in Chinese)

[20]YANG X, DING J. A computational framework for iceberg and ship discrimination: case study on Kaggle competition[J]. IEEE Access, 2020, 8: 82320-82327.

[21]ATAK U, ARSLANOGLU Y. Machine learning methods for predicting marine port accidents: a case study in container terminal[J]. Ships and Offshore Structures, 2022,17(11): 2480-2487.

[22]BENTEJAC C, CSORGO A. A comparative analysis of gradient boosting algorithms[J]. Artificial Intelligence Review, 2021, 54(3): 1937-1967.

[23]翁小雄,吕攀龙.基于GBDT算法的地铁IC卡通勤人群识别[J].重庆交通大学学报(自然科学版),2019,38(5)：8-12.

WENG X X, LV P L. Subway IC card commuter crowd identification based on GBDT algorithm[J]. Journal of Chongqing Jiaotong University (Natural Science), 2019, 38(5): 8-12. (in Chinese)

[24]谭广.基于E-LightGBM算法的5G套餐潜在客户识别研究[D].重庆:重庆工商大学,2022.

TAN G. Research on 5G package potential customer identification based on E-LightGBM algorithm[D]. Chongqing: Chongqing Technology and Business University, 2022. (in Chinese)

[25]赵娟娟,刘广臣,王瑞桃,等.基于LightGBM的风电机组齿轮箱油温故障预警研究[J].电力大数据,2021,24(11):76-84.

ZHAO J J, LIU G C, WANG R T, et al. Research on fault warning of wind turbine gearbox based on LightGBM algorithm model[J]. Power Systems and Big Data, 2021, 24(11): 76-84. (in Chinese)

[26]周挺,杨军,周强明,等.基于改进LightGBM的电力系统暂态稳定评估方法[J].电网技术,2019,43(6):1931-1940.

ZHOU T, YANG J, ZHOU Q M, et al. Power system transient stability assessment method based on modified LightGBM[J]. Power System Technology, 2019, 43(6): 1931-1940. (in Chinese)

[27]张亚伟,陈瑞凤,徐春婕,等.基于LSTM-LightGBM模型的车站环境温度预测[J].计算机测量与控制,2022,30(1):20-25.

ZHANG Y W, CHEN R F, XU C J, et al. Station environment temperature prediction based on LSTM-LightGBM model[J]. Computer Measurement & Control, 2022,30(1):20-25. (in Chinese)

[28]尹奇志,赵光普.船舶能效数据清洗方法研究[J].交通信息与安全,2017,35(3):68-73.

YIN Q Z, ZHAO G P. A study on data cleaning for energy efficiency of ships[J]. Journal of Transport Information and Safety, 2017, 35(3): 68-73. (in Chinese)

[29]王凯,徐浩,黄连忠,等．基于机器学习的船舶能耗智能预测方法分析[J].船舶工程,2020,42(11):87-93.

WANG K, XU H, HUANG L Z, et al. Analysis of intelligent prediction method of ship energy consumption based on machine learning[J]. Ship Engineering, 2020, 42(11):87-93. (in Chinese)

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献