基于激光雷达的散货堆场数字孪生建模方法

doi:10.16182/j.issn1004731x.joss.24-0461

摘要/Abstract

摘要：

由于散货码头大型设备多、作业环境恶劣、料堆形状时变等特点，导致在建立堆场模型时存在数据精度低、稳定性差等缺点，影响无人化、智能化作业控制。采用二维激光雷达结合设备机构运动扫描料堆点云数据，给出了一种用于散货堆场的数字孪生建模方法，包括堆场静态场景构建和料堆的实时建模。对于在港口堆场装卸作业中不改变表面形状的静态场景采用预制模型，而表面形状具有时变特征的料堆，设计了料堆的三维数据结构以及实时生成策略，采用料堆外形、截面、空间点的层次结构缓存数据，以最近距离截面作为策略条件实现料堆实时三维重建，满足自动化作业过程中孪生模型实时变化以及快速决策支持的要求。以某中转散货码头为案例，实现了该码头的散货堆场数字孪生建模，对系统的虚实一致性和实时性进行评估，验证了所述方法的可行性。

关键词: 数字孪生, 时变特征, 激光雷达, 三维重建, 实时生成策略

Abstract:

Due to the characteristics of large equipment, harsh working environment and time-varying shape of the material pile in bulk cargo terminal, there are some disadvantages such as low data accuracy and poor stability when building the storage yard model, which affects the unmanned and intelligent operation control. In this paper, we use two-dimensional laser radar combined with equipment mechanism motion to scan material pile point cloud data, present a digital twin modeling method for bulk storage yard, which includes static scene construction of storage yard and real-time modeling of material pile. Prefabricated models are used for the static scenes which do not change the surface shape during the loading and unloading operation of the material pile. For the pile with time-varying surface shape, the three-dimensional data structure and real-time generation strategy of the pile are designed. The hierarchical structure of the pile shape, section and space point is used to cache the data, and the nearest distance section is taken as the policy condition to realize real-time three-dimensional reconstruction of the pile, which meets the requirements of real-time changes of the twin model and fast decision support in the process of automatic operation. Finally, taking a bulk cargo terminal as an example, the digital twin modeling of the bulk cargo yard of the terminal is realized, and the consistency and real-time performance of the system are evaluated, verifying the feasibility of the method described in this paper.

Key words: digital twin, time-varying characteristic, 2D LiDAR, 3D reconstruction, real-time generation strategy

中图分类号:

U653.921

陆后军,朱益飞,荣延平等 . 基于激光雷达的散货堆场数字孪生建模方法[J]. 系统仿真学报, 2025, 37(9): 2269-2286.

Lu Houjun,Zhu Yifei,Rong Yanping,et al . Digital Twin Modeling Method for Bulk Cargo Stacks Based on 2D LiDAR[J]. Journal of System Simulation, 2025, 37(9): 2269-2286.

图/表 29

图1

图2

图3

图4

图5

图6

图7

图8

图9

表1

表2

不同货种建模参数

堆场区域	散货种类	l/m	w/m	$θ$ /(°)	h/m
7A	矿石	344	45	35	12
7B	矿石	382	69	35	12
7C	矿石	382	70	38	11
7D	矿石	382	49	38	11
8A	煤炭	275	40	37	10
8B	煤炭	174	40	37	10
9A	煤炭	345	31	37	10
9B	煤炭	365	49	37	10
9C	煤炭	335	35	37	10

表2

图10

表3

图11

表4

表5

图12

表6

图13

图14

表7

图15

图16

图17

表8

图18

图19

图20

图21

参考文献 24

[1]	张伟红, 周敬祥, 王荣明. 上海港罗泾散货码头管控一体化系统关键技术的研究及应用[C]//中国土木工程学会港口工程分会技术交流文集. 北京: 中国土木工程学会港口工程分会, 2009: 59-66.
[2]	刘世栋, 龚和艳, 王庚雨, 等. 干散货码头数字孪生系统研发及应用[J]. 水道港口, 2023, 44(4): 667-673.
	Liu Shidong, Gong Heyan, Wang Gengyu, et al. Research and Application of Digital Twin System for Dry Bulk Cargo Terminal[J]. Journal of Waterway and Harbor, 2023, 44(4): 667-673.
[3]	闫海龙. 干散货港口数字化堆场自动作业系统研究与设计[D]. 秦皇岛: 燕山大学, 2022.
	Yan Hailong. Research and Design of Automatic Operation System for Digital Storage Yard of Dry Bulk Port[D]. Qinhuangdao: Yanshan University, 2022.
[4]	曹帅. 散料料堆的三维重建与优化研究[D]. 秦皇岛: 燕山大学, 2022.
	Cao Shuai. Research on 3D Reconstruction and Optimization of Bulk Material Pile[D]. Qinhuangdao: Yanshan University, 2022.
[5]	Cheng Liang, Wu Yang, Wang Yu, et al. Three-dimensional Reconstruction of Large Multilayer Interchange Bridge Using Airborne LiDAR Data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(2): 691-708.
[6]	Li Zimo, Gogia P C, Kaess M. Dense Surface Reconstruction from Monocular Vision and LiDAR[C]//2019 International Conference on Robotics and Automation (ICRA). Piscataway: IEEE, 2019: 6905-6911.
[7]	Zhang Yang, Liu Wei, Wang Fuji, et al. Improved Separated-parameter Calibration Method for Binocular Vision Measurements with a Large Field of View[J]. Optics Express, 2020, 28(3): 2956-2974.
[8]	Hasch Jürgen, Topak Eray, Schnabel Raik, et al. Millimeter-wave Technology for Automotive Radar Sensors in the 77 GHz Frequency Band[J]. IEEE Transactions on Microwave Theory and Techniques, 2012, 60(3): 845-860.
[9]	陈建辉, 朱方文, 袁政鹏. 激光传感器料堆扫描的三维仿真[J]. 机电一体化, 2008, 14(7): 56-58, 64.
	Chen Jianhui, Zhu Fangwen, Yuan Zhengpeng. 3D Simulation of Scanning Piles by Laser Sensor[J]. Mechatronics, 2008, 14(7): 56-58, 64.
[10]	张荠匀, 王建军, 李旭辉, 等. 基于激光雷达的料堆特征提取方法优化[J]. 激光与光电子学进展, 2023, 60(2): 364-370.
	Zhang Jiyun, Wang Jianjun, Li Xuhui, et al. Optimization of Feature-extraction Method for Stockpiled Materials Based on LiDAR[J]. Laser & Optoelectronics Progress, 2023, 60(2): 364-370.
[11]	张永平, 左颖, 刘博, 等. 数字孪生车间制造运营管理平台[J]. 计算机集成制造系统, 2024, 30(1): 1-13.
	Zhang Yongping, Zuo Ying, Liu Bo, et al. Digital Twin Shop-floor Manufacturing Operations Management Platform[J]. Computer Integrated Manufacturing Systems, 2024, 30(1): 1-13.
[12]	刘新宇, 韩旭, 王泓晖, 等. 高速列车悬挂系统的数字孪生模型自适应实时修正方法[J/OL]. 计算机集成制造系统. (2023-11-01) [2024-04-28]. .
	Liu Xinyu, Han Xu, Wang Honghui, et al. Adaptive Real-time Correction Method for Digital Twin Model of High-speed Train Suspension System[J/OL]. Computer Integrated Manufacturing Systems. (2023-11-01) [2024-04-28]. .
[13]	孙贺, 夏唐斌, 石易达, 等. 基于数字孪生的返工型生产系统能耗在线优化[J]. 计算机集成制造系统, 2023, 29(1): 11-24.
	Sun He, Xia Tangbin, Shi Yida, et al. Online Energy Consumption Optimization of Rework Production System Based on Digital Twin[J]. Computer Integrated Manufacturing Systems, 2023, 29(1): 11-24.
[14]	缪天越, 王璐, 何家孝, 等. 融合强化学习的工业机器人数字孪生仿真方法研究[J]. 系统仿真学报, 2024, 36(12): 2971-2983.
	Miao Tianyue, Wang Lu, He Jiaxiao, et al. Research on Digital Twin Simulation Method of Industrial Robot Integrated with Reinforcement Learning[J]. Journal of System Simulation, 2024, 36(12): 2971-2983.
[15]	王红军, 林俊强, 邹湘军, 等. 基于数字孪生的果园虚拟交互系统构建[J]. 系统仿真学报, 2024, 36(6): 1493-1508.
	Wang Hongjun, Lin Junqiang, Zou Xiangjun, et al. Construction of a Virtual Interactive System for Orchards Based on Digital Twin[J]. Journal of System Simulation, 2024, 36(6): 1493-1508.
[16]	吕楠, 王琪冰, 陆佳炜, 等. 基于数字孪生的自动扶梯乘客行为监测方法研究[J]. 系统仿真学报, 2023, 35(8): 1737-1747.
	Nan Lü, Wang Qibing, Lu Jiawei, et al. Monitoring Method Research on Passenger Behavior on Escalator Based on Digital Twin[J]. Journal of System Simulation, 2023, 35(8): 1737-1747.
[17]	张在房, 孙金, 高楠, 等. 基于数字孪生的肺结节诊断[J]. 计算机集成制造系统, 2023, 29(6): 1894-1904.
	Zhang Zaifang, Sun Jin, Gao Nan, et al. Diagnosis of Pulmonary Nodules Using Digital Twins[J]. Computer Integrated Manufacturing Systems, 2023, 29(6): 1894-1904.
[18]	任乾坤, 熊鑫立, 刘京菊, 等. 网络空间安全中的数字孪生技术研究[J]. 系统仿真学报, 2024, 36(8): 1944-1957.
	Ren Qiankun, Xiong Xinli, Liu Jingju, et al. Reserach on Digital Twins Technology in Cyberspace Security[J]. Journal of System Simulation, 2024, 36(8): 1944-1957.
[19]	陈培, 刘超, 蔡黄河. 基于Unity 3D的集装箱码头数字孪生系统设计[J]. 港口科技, 2021(7): 8-10, 42.
[20]	李玉, 苌道方, 高银萍, 等. 基于数字孪生的自动化集装箱码头多AGV动态调度[J]. 计算机集成制造系统, 2023, 29(12): 4175-4190.
	Li Yu, Chang Daofang, Gao Yinping, et al. Multi-AGV Dynamic Scheduling of Automated Container Terminal Based on Digital Twin[J]. Computer Integrated Manufacturing Systems, 2023, 29(12): 4175-4190.
[21]	罗威强, 王米换. 基于数字孪生技术的散货码头料场管理系统[J]. 港口科技, 2021(12): 10-15.
[22]	陈士杰, 张森林, 刘妹琴, 等. 基于改进Delaunay三角剖分的水下地形三维重建算法[J]. 计算机科学, 2020, 47(11): 137-141.
	Chen Shijie, Zhang Senlin, Liu Meiqin, et al. Underwater Terrain Three-dimensional Reconstruction Algorithm Based on Improved Delaunay Triangulation[J]. Computer Science, 2020, 47(11): 137-141.
[23]	张望, 李瑛, 杨德山. 基于三维激光扫描的散货堆快速建模[J]. 应用激光, 2019, 39(2): 340-345.
	Zhang Wang, Li Ying, Yang Deshan. Rapid Modeling of Stockpile Based on 3D Laser Scanning[J]. Applied Laser, 2019, 39(2): 340-345.
[24]	申禹, 李志伟. 大型干散货港口堆取料机自动化作业系统的研究与应用[J]. 自动化应用, 2024, 65(10): 14-16.
	Shen Yu, Li Zhiwei. Research and Application of Automatic Operation System of Stacker-reclaimer in Large Dry Bulk Port[J]. Automation Application, 2024, 65(10): 14-16.

参数	数值	参数	数值
大车速度/(m/min)	4～30	h₂/m	10
测量角度范围/(°)	0～180	h₁/m	2
角度分辨率/(°)	0.5	α/(°)	-12～+6
单次扫描周期/ms	26.64	β/(°)	-120～+120

模型类型	模型清单
整体模型	港口整体模型，地形分割(包括堆场、引桥、泊位、绿化区、办公区)
静态模型	行政楼、引桥建筑、变电所
船只模型	10种以上船只模型，舱口需要单独建模，船用起重机
设备	装船机：含装船机主要零件部位卸船机：含卸船机主要零件部位皮带机：廊道、廊道内皮带、皮带驱动、转运站内部设施斗轮机：含斗轮机主要零件部位
堆场	料堆、轨道、防尘网、挡矿墙
其他	路灯、道路、监控摄像头、照明灯、消防栓、绿化带、路牌、指示标志

项目			物理实体	孪生体
建筑	行政楼高度/m		14.4	14.4
建筑	行政楼区域面积/m²		386.1	386.2
船只	船只长度/m		177.6	177.2
	舱口面积/m²		120.5	120.5
	船舶高度/m		13.5	13.5
特定堆区	料堆高度/m		10.7	10.6
	堆区面积/m²		10 695.0	10 693.2
	安歇角/(°)		34.4	33.7
设备	廊道长度/m		337.7	337.5
	廊道高度/m		8.4	8.4
	斗轮机尺寸/m	高度	21.7	21.7
		大臂	30.0	30.0
		斗轮半径	6.7	6.7

方案序号	о₁	о₂	о₃
S1	30	0.4	0.4
S2	31	0.4	0.4
S3	30	0.5	0.5
S4	31	0.5	0.5
S5	30	0.6	0.6
S6	31	0.6	0.6

序号	实际高度	S1方案		S2方案		S3方案		S4方案		S5方案		S6方案
序号	实际高度	高度	误差	高度	误差	高度	误差	高度	误差	高度	误差	高度	误差

1	30	46	-16	40	-10	38	-8	34	-4	35	-5	38	-8
2	90	109	-19	97	-7	96	-6	96	-6	99	-9	76	14
3	0	-11	11	6	-6	-3	3	1	-1	-6	6	12	-12
4	531	510	21	523	8	548	-17	544	-13	548	-17	546	-15
5	912	889	23	926	-14	903	9	921	-9	926	-14	926	-14
6	1 146	1 154	-8	1 158	-12	1 162	-16	1 143	3	1 157	-11	1 153	-7
7	1 186	1 164	22	1 171	15	1 197	-11	1 177	9	1 176	10	1 167	19
8	1 173	1 199	-26	1 162	11	1 163	10	1 183	-10	1 182	-9	1 184	-11
9	1 059	1 066	-7	1 078	-19	1 058	1	1 054	5	1 063	-4	1 064	-5
10	1 169	1 178	-9	1 153	16	1 165	4	1 163	6	1 187	-18	1 173	-4
11	1 084	1 069	15	1 094	-10	1 074	10	1 088	-4	1 064	20	1 095	-11
12	851	864	-13	843	8	843	8	849	2	861	-10	864	-13
13	746	771	-25	763	-17	759	-13	763	-17	734	12	749	-3
14	681	683	-2	668	13	674	7	679	2	693	-12	689	-8
15	26	22	4	21	5	21	5	22	4	19	7	19	7
16	0	13	-13	-3	3	-6	6	0	0	6	-6	-4	4
17	639	657	-18	647	-8	647	-8	643	-4	648	-9	645	-6

18	607	614	-7	631	-24	621	-14	614	-7	621	-14	613	-6
19	1 066	1 056	10	1 078	-12	1 074	-8	1 064	2	1 071	-5	1 074	-8
20	1 095	1 076	19	1 106	-11	1 102	-7	1 102	-7	1 085	10	1 106	-11
21	1 146	1 163	-17	1 152	-6	1 132	14	1 151	-5	1 154	-8	1 152	-6
22	1 184	1 174	10	1 196	-12	1 192	-8	1 189	-5	1 179	5	1 196	-12
23	1 086	1 061	25	1 059	27	1 072	14	1 074	12	1 098	-12	1 089	-3
24	1 083	1 104	-21	1 068	15	1 081	2	1 096	-13	1 074	9	1 074	9
25	961	957	4	977	-16	956	5	959	2	969	-8	951	10
26	143	152	-9	167	-24	139	4	148	-5	158	-15	151	-8
27	39	43	-4	31	8	42	-3	36	3	31	8	46	-7
平均误差		14.0		12.5		8.2		5.9		10.1		8.9