面向能效优化的内河航道智能划分方法

张瀚宇; 尹奇志; 王春英; 钱巍文; 张露; 秦乐天

doi:10.3963/j.jssn.1674-4861.2025.01.009

面向能效优化的内河航道智能划分方法

doi: 10.3963/j.jssn.1674-4861.2025.01.009

张瀚宇^{1, 2,},
尹奇志^{2, 3, 4, ,},
王春英⁵,
钱巍文^{2, 3, 4},
张露⁵,
秦乐天^{2, 3}

1.
武汉理工大学船海与能源动力工程学院武汉 430063
2.
武汉理工大学水路交通控制全国重点实验室武汉 430063
3.
武汉理工大学交通与物流工程学院武汉 430063
4.
湖北东湖实验室武汉 420202
5.
潍柴动力股份有限公司山东潍坊 261061

基金项目:

国家重点研发计划项目 2022YFB4300803

工信部高技术船舶科研项目 MC-202002-C03

潍柴动力股份有限公司技术项目 WCDL-GH-2021-0050

详细信息

作者简介:
张瀚宇（1998—），硕士研究生. 研究方向：船舶能效建模与优化. E-mail: hyzhang@whut.edu.cn

通讯作者:
尹奇志（1976—），博士，副教授. 研究方向：船舶能效提升技术. E-mail: qzyin@whut.edu.cn

中图分类号: U676.3
计量
- 文章访问数: 17
- HTML全文浏览量: 8
- PDF下载量: 0
- 被引次数: 0
出版历程
- 收稿日期: 2024-05-13
- 网络出版日期: 2025-06-27

Intelligent Segmentation Method of Inland Waterway for Energy Efficiency Optimization

ZHANG Hanyu^{1, 2
,},
YIN Qinzhi^{2, 3, 4
, ,},
JIN Liangzhen⁵,
QIAN Weiwen^{2, 3, 4},
ZHANG Wenqiang⁵,
QIN Letian^{2, 3}

1.
School of Naval Architecture Ocean and Energy Power Engineering, Wuhan University of Technology, Wuhan 430063, China
2.
State Key Laboratory of Maritime Technology and Safety, Wuhan University of Technology, Wuhan 430063, China
3.
School of Transportation and Logistics Engineering, Wuhan University of Technology, Wuhan 430063, China
4.
East Lake Laboratory, Wuhan 420202, China
5.
Weichai Power Company Limited, Weifang 261061, Shandong, China

摘要

摘要: 内河航道航段的自动科学划分对提升船舶能效模型的精度具有重要意义。针对船舶能效优化过程中，根据航段划分结果构建的各航段油耗预测模型和航速预测模型精度不高等问题，研究了1种基于能效优化的内河航道智能划分方法。该方法通过归一化处理以及构建通航环境参数与船舶能效的相关性系数将通航环境参数对船舶能效的影响程度纳入到通航环境数据聚类中，运用K-means聚类算法将整条航线划分为多段；运用随机森林（random forest，RF）算法建立各航段的船舶油耗预测模型和船舶航速预测模型；通过优化聚类数量实现船舶能效模型综合平均绝对百分比误差（mean absolute percentage error，MAPE）最低。以1艘内河散货船为研究对象，对该方法进行实例应用和验证，并分析了数据量对船舶能效模型综合精度的影响。研究结果表明：通过优化聚类数量可以实现船舶能效模型综合精度的提升，采用该方法构建第1航次的油耗预测模型和航速预测模型，实现船舶能效模型综合MAPE由3.53%降低至3.32%；增加构建船舶能效模型的能效相关数据的数据量有利于船舶能效模型综合精度的提升，当对象船舶用于建模的能效数据由1个航次增加到5个航次时，船舶能效模型综合MAPE由3.32%降低至1.65%；不同建模数据的最佳聚类数量不同，通过取多航次的综合最佳聚类数可实现航道的优化划分；所提出航道划分方法得到的能效模型综合MAPE比常用航道划分方法得到的能效模型综合MAPE降低了0.54%，验证了所提出的航道划分方法对提升船舶能效预测模型精度的有效性。
- 能效优化 /
- 内河船舶 /
- 航段划分 /
- K-means聚类算法
Abstract: The automatic scientific segmentation of inland waterways is significant to enhancing the accuracy of ship energy efficiency models. To address the problem of low accuracy in fuel consumption and speed prediction models built based on conventional waterway segmentation for energy efficiency optimization, an intelligent segmentation method of inland waterways aimed at optimizing energy efficiency is studied. The method incorporates the influence of navigational environmental parameters on ship energy performance into the clustering process by normalizing the data and calculating correlation coefficients between environmental parameters and energy efficiency indicators. The K-means clustering algorithm is employed to segment the entire route into multiple sections. For each segment, Random Forest algorithm is used to establish models for fuel consumption and ship speed prediction. The optimal number of clusters is determined by minimizing the overall mean absolute percentage error (MAPE) of the energy efficiency models. A case study involving an inland bulk carrier is conducted to demonstrate the application and validate the effectiveness of the proposed method, along with an analysis of the influence of data volume on the model accuracy. The results show that optimizing the number of clusters significantly improves model accuracy: in the firstly voyage, the comprehensive MAPE of the energy efficiency model is reduced from 3.53% to 3.32%. Increasing the volume of energy-related data used in model construction further enhances the performance: when the dataset expanded from one voyage to five, the comprehensive MAPE decreased from 3.32% to 1.65%. The optimal number of clusters varies with different datasets, and selecting an optimal number of clusters based on multi-voyage data leads to the optimal waterway segmentation. Compared to the conventional segmentation method, the proposed approach reduced the comprehensive MAPE of energy efficiency model by 0.54%, validating the effectiveness of the method in improving the prediction accuracy of ship energy efficiency model.
- energy efficiency optimization /
- inland ships /
- waterway segmentation /
- K-means clustering algorithm

HTML全文

图 1 基于能效优化的内河航道智能划分方法具体流程图

Figure 1. Specific flowchart of intelligent division method of inland waterways based on energy efficiency optimization

下载: 全尺寸图片幻灯片

图 2 对象船舶能效采集系统

Figure 2. Energy efficiency acquisition system of the target ship

下载: 全尺寸图片幻灯片

图 3 各航次不同聚类数量的MAPE对比

Figure 3. Comparison of MAPE with different number of clusters in the differemt cruise

下载: 全尺寸图片幻灯片

图 4 第1航次n=4的航段划分结果示意图

Figure 4. Schematic diagram of segment division result for the first cruise with n=4

下载: 全尺寸图片幻灯片

图 5 第1航次的各航段能效模型MAPE示意图

Figure 5. Schematic diagram of the energy efficiency model MAPE for each segment of the first cruise

下载: 全尺寸图片幻灯片

图 6 不同数据量的MAPE对比

Figure 6. Comparison of MAPE with different data volumes

下载: 全尺寸图片幻灯片

图 7 不同聚类数量的平均MAPE对比

Figure 7. Comparison of average MAPE with different number of clusters

下载: 全尺寸图片幻灯片

图 8 第1~5航次n=10的航段划分结果示意图

Figure 8. Schematic diagram of the results of the segment division of cruises 1 to 5 with n=10

下载: 全尺寸图片幻灯片

表 1 对象船舶主要参数

Table 1. Main parameters of the target ship

参数	数值
船长/m	99.8
型宽/m	16.2
型深/m	5.8
主机功率/kw	2×850
吃水/m	4.9
排水量/t	6 949.4

下载: 导出CSV

表 2 通航环境与能效之间的相关系数

Table 2. Correlation coefficients between navigational environment and energy efficiency

能效参数	通航环境
能效参数	风速	风向	水深	水流速度
油耗	0.023	0.034	-0.346	-0.275
航速	0.33	-0.271	0.029	0.741

下载: 导出CSV

表 3 预处理后的部分通航环境数据

Table 3. Partial avigational environment data after preprocessing

风速	风向	水深	水流速度
0.062 5	-0.032 1	-0.014 1	0.146 5
0.059 8	-0.039 6	-0.014 8	0.147 6
0.055 6	-0.105 3	-0.013 9	0.146 9
0.040 3	-0.105 3	-0.086 4	0.127 1
0.043 1	-0.037 3	-0.085 5	0.135 3
⋮	⋮	⋮	⋮

下载: 导出CSV

表 4 RF算法的超参数设置

Table 4. Hyperparameter settings for RF algorithm

模型	超参数类型	不同航次参数值
模型	超参数类型	第1航次	第1~2航次	第1~3航次	第1~4航次	第1~5航次
油耗预测模型	n_estimators	104	92	94	118	180
	max_depth	22	24	25	26	25
	min_samples_leaf	2	4	2	3	4
	min_samples_split	2	6	14	14	6
	max_leaf_nodes	3 382	3 979	3 743	3 752	3 990
航速预测模型	n_estimators	104	110	113	116	175
	max_depth	19	23	23	26	26
	min_samples_leaf	1	3	3	3	3
	min_samples_split	12	3	14	5	15
	max_leaf_nodes	3 758	3 990	3 972	3 944	3 995

下载: 导出CSV

表 5 划分方法对能效模型精度的影响

Table 5. The impact of segment division methods on the accuracy of energy efficiency models

划分方法	最佳聚类数量n	油耗预测模型MAPE/%	油耗预测模型MAPE/%	能效模型综合MAPE/%
常用方法	5	2.14	2.23	2.19
本文方法	10	1.58	1.72	1.65

下载: 导出CSV

参考文献(22)

[1]	YUAN J, NIAN V, HE J L, et al. Cost-effectiveness analysis of energy efficiency measures for maritime shipping using a metamodel based approach with different data sources[J]. Energy, 2019, 189(15): 116205。
[2]	严新平, 刘佳仑, 范爱龙, 等. 智能船舶技术发展与趋势简述[J]. 船舶工程, 2020, 42(3): 15-20. YAN X P, LIU J L, FAN A L, et al. Introduction to the development and trends of intelligent ship technology[J]. Ship Engineering, 2020, 42(3): 15-20. (in Chinese)
[3]	FAN A L, YAN X P, RICHARD B, et al. A novel ship energy efficiency model considering random environmental parameters[J]. Journal of Marine Engineering & Technology, 2020, 19: 215-228.
[4]	陈兴. 内河船舶不同营运阶段能效优化方法研究[D]. 武汉: 武汉理工大学, 2021. CHEN X. Study on energy efficiency optimization methods of inland ships in different operation stages[D]. Wuhan: Wuhan University of Technology, 2021. (in Chinese)
[5]	YAN X P, WANG K, YUAN Y P, et al. Energy-efficient shipping: an application of big data analysis for optimizing engine speed of inland ships considering multiple environmental factors[J]. Ocean Engineering, 2018, 169: 457-468. doi: 10.1016/j.oceaneng.2018.08.050
[6]	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)
[7]	霍得利. 船舶航速优化节能性研究[D]. 大连: 大连海事大学, 2017. HUO D L. Study on energy conservation of ship speed optimization[D]. Dalian: Dalian Maritime University, 2017. (in Chinese)
[8]	WANG K, YAN X P, YUAN Y P, et al. Study on route division for ship energy efficiency optimization based on big environment data[C]. 4^th International Conference on Transportation Information and Safety, Banff, Canada: IEEE, 2017.
[9]	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(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]. Ships, 2018, 29(1): 8-17. (in Chinese)
[10]	王寰宇. 远洋船舶分段航速优化及其智能算法研究[D]. 大连: 大连海事大学, 2018. WANG H Y. Ocean ship segmentation speed optimization and intelligent algorithms[D]. Dalian: Dalian Maritime University, 2018. (in Chinese)
[11]	袁智. 内河水域船舶运行数据分析与能效优化方法研究[D]. 武汉: 武汉理工大学, 2021. YUAN Z. Operational data analysis and energy efficiency optimization method for inland river ship[D]. Wuhan: Wuhan University of Technology, 2021. (in Chinese)
[12]	陈雪梅. 基于数据挖掘的内河船舶航速优化研究[D]. 重庆: 重庆交通大学, 2022. CHEN X M. Research on speed optimization of inland ships based on data mining[D]. Chongqing: Chongqing Jiaotong University, 2022. (in Chinese)
[13]	王壮, 王凯, 黄连忠, 等. 海况识别下的船舶航速动态优化方法[J]. 哈尔滨工程大学学报, 2022, 43(4): 488-494. WANG Z, WANG K, HUANG L Z, et al. Dynamic optimization method of ship speed based on sea condition recognition[J]. Journal of Harbin Engineering University, 2022, 43 (4): 488-494. (in Chinese)
[14]	MA L, YANG P, GAO D J, et al. A multi-objective energy efficiency optimization method of ship under different sea conditions[J]. Ocean Engineering, 2023, 290: 116337.
[15]	陈佳良, 胡钊政, 李飞. 基于时空特征序列匹配的交通流状态估计方法[J]. 交通信息与安全, 2021, 39(3): 68-76, 120. doi: 10.3963/j.jssn.1674-4861.2021.03.009 CHEN J L, HU Z Z, LI F. An estimation method of traffic flow state based on matching of temporal-spatial feature sequences[J]. Journal of Transport Information and Safety, 2021, 39(3): 68-76, 120. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2021.03.009
[16]	宋嘎, 王娜娜. 坐底式深水网箱振动实测信号分析处理方法[J]. 船海工程, 2020, 49(3): 117-120. SONG G, WANG N N. Analysis and processing method of vibration measured signal of bottom-deep-water cage[J]. Ship&Ocean Engineering, 2020, 49(3): 117-120. (in Chinese)
[17]	王康力. 面向航速与纵倾优化的船舶油耗模型建模方法研究[D]. 天津: 天津理工大学, 2024. WANG K L. Research on ship fuel consumption modelling approaches for speed and trim optimization[D]. Tianjin: Tianjin University of Technology, 2024. (in Chinese)
[18]	朱晓晨, 尹奇志, 赵福芹, 等. 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. ZHU X C, YIN Q Z, ZHAO F Q, et al. Ship speed prediction model based on LightGBM[J]. Journal of Dalian Maritime University, 2023, 49(1): 56-65. (in Chinese)
[19]	周田瑞. 基于监测数据的船舶能耗建模及航速优化研究[D]. 上海: 上海海事大学, 2023. ZHOU T R. Research on fuel consumption modeling and speed optimization based on real ship monitoring data[D]. Shanghai: Shanghai Maritime University, 2023. (in Chinese)
[20]	赵世博. 内河船舶能效监测、分析与评估方法研究[D]. 武汉: 武汉理工大学, 2022. ZHAO S B. Study on energy efficiency monitoring, analysis and assessment methods of inland ships[D]. Wuhan: Wuhan University of Technology, 2022. (in Chinese)
[21]	陈帅, 蒋彩霞, 王子渊, 等. 基于AutoML的低频波浪载荷智能预报方法[J]. 船舶工程, 2023, 45(7): 124-131, 142. CHEN S, JIANG C X, WANG Z Y. Intelligent prediction method of low frequency wave load based on AutoML[J]. Ship Engineering, 2023, 45(7): 124-131, 142. (in Chinese)
[22]	王静. 基于GPS数据的货运车辆出行特性研究[D]. 北京: 北京交通大学, 2022. WANG J. Research on travel characteristics of freight vehicles based on GPS data[D]. Beijing: Beijing Jiaotong University, 2022. (in Chinese)