基于图神经网络的路面病害态势预测方法

doi:10.3785/j.issn.1008-973X.2024.12.019

浙江大学学报(工学版)

2024, Vol. 58

Issue (12): 2596-2608 DOI: 10.3785/j.issn.1008-973X.2024.12.019

交通工程

基于图神经网络的路面病害态势预测方法

马泽超1(

),刘小明1,*(

),夏汗青2,王伟强1,王久增3,申海涛3

1. 北方工业大学电气与控制工程学院，北京 100144
2. 南京航空航天大学民航学院，江苏南京 211106
3. 唐山高速公路集团有限公司，河北唐山 063000

Pavement distress situation prediction method based on graph neural network

Zechao MA1(

),Xiaoming LIU1,*(

),Hanqing XIA2,Weiqiang WANG1,Jiuzeng WANG3,Haitao SHEN3

1. School of Electrical and Control Engineering, North China University of Technology, Beijing 100144, China
2. College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
3. Tangshan Expressway Group Company, Tangshan 063000, China

全文: PDF(1111 KB) HTML

摘要：

针对路面病害生成和恶化的预测问题，提出应用图卷积神经网络的路面病害态势预测方法. 通过聚类算法建立拓扑网络，选取目标病害在演化过程中的主要影响因素；为了增强图神经网络对病害信息的表达能力，采用图拓扑增强的方法，从静态和动态方面分别构造与病害信息相关的视图；采用图神经网络(GNN)架构增强的方法，在视图维度上应用注意力机制调整不同视图的影响力，并在时间维度上应用Transformer和GRU模块，增强模型在长时间序列中对病害状态的预测性能. 设计模型的内部调整测试，经消融试验、多样本测试和超参数对照组的验证，证明所提模型的适用性和稳定性. 针对大型稀疏的路面病害数据集，此模型的平均绝对误差均值收敛在4.0以内，综合性能优于传统预测算法.

关键词： 公路养护; 路面病害; 图神经网络; 时间序列预测; 裂缝

Abstract:

A road pavement distress situation forecasting method employing graph convolutional networks was introduced, addressing the prediction problem of road pavement distress generation and deterioration. Firstly, a topological network was established through clustering algorithms, selecting the main influencing factors of the target pavement distress during its evolution. Subsequently, to enhance the expressive capability of the graph neural network for distress information, a graph topology enhancement method was employed, constructing views related to distress information from both static and dynamic aspects. Finally, an enhanced graph neural network (GNN) architecture was applied, by incorporating attention mechanisms in the view dimension to adjust the influence of different views and utilizing Transformer and GRU modules in the temporal dimension to enhance the predictive performance of the model for pavement distress states over extended time sequences. The internal calibration tests of the model, including ablation studies, multi-sample testing, and hyperparameter control group validation, demonstrated the applicability and stability of the proposed model. For the large and sparse pavement disease dataset, the mean absolute error of this model converged within 4.0, which was better than the results of the traditional prediction algorithms in terms of comprehensive performance.

Key words: highway maintenance pavement distress graph neural network time series forecast crack

收稿日期: 2023-10-23 出版日期: 2024-11-25

CLC:

F 287.3

基金资助: 北方工业大学毓秀创新项目 (2024NCUTYXCX109).

通讯作者: 刘小明 E-mail: z2mazc@126.com;tslxm@sina.com

作者简介: 马泽超（2000—），男，硕士生，从事智能交通、交通时空大数据研究. orcid.org/0009-0009-2552-6280. E-mail：z2mazc@126.com

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	作者相关文章
	马泽超
	刘小明
	夏汗青
	王伟强
	王久增
	申海涛

引用本文:

马泽超,刘小明,夏汗青,王伟强,王久增,申海涛. 基于图神经网络的路面病害态势预测方法[J]. 浙江大学学报(工学版), 2024, 58(12): 2596-2608.

Zechao MA,Xiaoming LIU,Hanqing XIA,Weiqiang WANG,Jiuzeng WANG,Haitao SHEN. Pavement distress situation prediction method based on graph neural network. Journal of ZheJiang University (Engineering Science), 2024, 58(12): 2596-2608.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2024.12.019 或 https://www.zjujournals.com/eng/CN/Y2024/V58/I12/2596

图 1 图卷积神经网络结构

图 2 佛罗里达州各类型路面病害的发生频率

图 3 佛罗里达州部分路面病害之间的线性拟合图

图 4 佛罗里达州的病害相关性热力图

图 5 多视图图卷积模型结构

图 6 多视图案例的构建

表 1 MV-GCN与基线模型的平均性能对比

图 7 MV-GCN与基线模型的逐月预测性能对比

表 2 视图对性能的影响

表 3 多重注意力模块对性能的影响

表 4 基于佛罗里达州的聚类簇的预测结果

图 8 路面病害的预测值和真实值的对比

表 5 不同地区不同类型的路面病害的预测结果

表 6 改变GCN层数的性能对比结果

表 7 不同训练比下的性能比较结果

表 8 不同相似度算法的性能对比结果

1	PAN Y, SHANG Y, LIU G, et al Cost-effectiveness evaluation of pavement maintenance treatments using multiple regression and life-cycle cost analysis[J]. Construction and Building Materials, 2021, 292 (6): 123461
2	HAFEZ M, KSAIBATI K, ATADERO R Developing a methodology to evaluate the effectiveness of pavement treatments applied to low-volume paved roads[J]. International Journal of Pavement Engineering, 2019, 20 (8): 894- 904
3	FRANCE-MENSAH J, KOTHARI C, BRIEN W J, et al Integrating social equity in highway maintenance and rehabilitation programming: a quantitative approach[J]. Sustainable Cities and Society, 2019, 48 (11): 101526
4	MARCELINO P, ANTUNES M L, FORTUNATO E, et al Machine learning approach for pavement performance prediction[J]. International Journal of Pavement Engineering, 2021, 22 (3): 341- 354
5	何西西. 基于大数据分析的沥青路面使用性能指标评价研究[D]. 南京: 东南大学, 2020. HE Xixi. Research on the evaluation of asphalt pavement performance index based on big data analysis [D]. Nanjing: Southeast University, 2020.
6	MARCELINO P, ANTUNES M L, FORTUNATO E, et al Transfer learning for pavement performance prediction[J]. International Journal of Pavement Research and Technology, 2020, 13 (3): 154- 167
7	PITYONESI S M, EL-DIRABY T E Examining the relationship between two road performance indicators: pavement condition index and international roughness index[J]. Transportation Geotechnics, 2020, 26: 100441
8	邬晓光, 时元绪, 李院军, 等基于灰色理论的高速公路小修保养维修量预测模型[J]. 公路工程, 2020, 45: 123- 127 WU Xiaoguang, SHI Yuanxu, LI Yuanjun, et al Prediction model of minor repair and maintenance amount of expressway based on grey theory[J]. Highway Engineering, 2020, 45: 123- 127
9	史小丽, 王峰, 张平, 等. 高速公路小修保养工程量预估方法[C]//中国公路学会养护与管理分会第八届学术年会论文集. 厦门: 中国公路学会, 2018. SHI Xiaoli, WANG Feng, ZHANG Ping, et al. Prediction method of highway minor repair and maintenance works [C]// Proceedings of 8th Annual Conference of Maintenance and Management Branch of China Highway Society . Xiamen: China Highway, 2018.
10	ABDELAZIZ N, EL-HAKIM R T, EL-BADAWY S M, et al International roughness index prediction model for flexible pavements[J]. International Journal of Pavement Engineering, 2020, 21 (1): 88- 99 doi: 10.1080/10298436.2018.1441414
11	滕伟玲, 姚玉玲高速公路小修保养工程量的预测模型[J]. 长安大学学报: 自然科学版, 2012, 32 (6): 23- 27 TENG Weiling, YAO Yuling Expressway minor maintenance amount prediction based on neural network[J]. Journal of Chang’an University: Natural Science Edition, 2012, 32 (6): 23- 27
12	NYIRANDAYISABYE R, LI H, DONG Q, et al Automatic pavement damage predictions using various machine learning algorithms: evaluation and comparison[J]. Results in Engineering, 2022, 16: 100657
13	PANDIT W H, SHARMA K P, SHARMA N. International roughness index (IRI) prediction using various machine learning techniques on Flexible Pavements [C]// 2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering . Greater Noida: IEEE, 2022: 1899–1905.
14	INKOOM S, SOBANJO J, BARBU A, et al Prediction of the crack condition of highway pavements using machine learning models[J]. Structure and Infrastructure Engineering, 2019, 15 (7): 940- 953
15	CHOPRA T, PARIDA M, KWATRA N, et al Development of pavement distress deterioration prediction models for urban road network using genetic programming[J]. Advances in Civil Engineering, 2018, (2): 1253108
16	LI J, YIN G, WANG X, et al Automated decision making in highway pavement preventive maintenance based on deep learning[J]. Automation in Construction, 2022, 135 (6): 104111
17	KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks [C]// 2017 International Conference on Learning Representations . Toulon: OpenReview, 2017.
18	LI Y, YU R, SHAHABI C, et al. Graph convolutional recurrent neural network: data-driven traffic forecasting [C]// 2018 International Conference on Learning Representations . Vancouver: OpenReview, 2018.
19	LI M, ZHU Z. Spatial-temporal fusion graph neural networks for traffic flow forecasting [C]// Proceedings of the AAAI Conference on Artificial Intelligence . Washington: AAAI, 2021: 4189–4196.
20	GUO K, HU Y, QIAN Z, et al Optimized graph convolution recurrent neural network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22 (2): 1138- 1149
21	SHI X, QI H, SHEN Y, et al A spatial-temporal attention approach for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22 (8): 4909- 4918
22	TRIRAT P, YOON S, LEE J MG-TAR: multi-view graph convolutional networks for traffic accident risk prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24 (4): 3779- 3794
23	GUO J, SONG C, WANG H. A multi-step traffic speed forecasting model based on graph convolutional LSTM [C]// 2019 Chinese Automation Congress . Hangzhou: IEEE, 2019: 2466–2471.
24	ZHAO L, SONG Y, ZHANG C, et al T-GCN: a temporal graph convolutional network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21 (9): 3848- 3858
25	周毅, 胡姝婷, 李伟, 等图神经网络驱动的交通预测技术: 探索与挑战[J]. 物联网学报, 2021, 5 (4): 1- 16 ZHOU Yi, HU Shuting, LI Wei, et al Graph neural network driven traffic prediction technology: review and challenge[J]. Chinese Journal on Internet of Things, 2021, 5 (4): 1- 16 doi: 10.11959/j.issn.2096-3750.2021.00235
26	柴明堂, 马腾, 李国玉, 等新藏公路路面病害空间分布及相互关系分析[J]. 冰川冻土, 2022, 44 (6): 1681- 1693 CHAI Mingtang, MA Teng, LI Guoyu, et al Analysis on spatial distribution and correlation of pavement distress along the Xinjiang-Tibet highway[J]. Journal of Glaciology and Geocryology, 2022, 44 (6): 1681- 1693
27	PHILIP B, JASSMI H A A bayesian approach towards modelling the interrelationships of pavement deterioration factors[J]. Buildings, 2022, 12 (7): 1039
28	易辉. 高速公路沥青路面使用性能评价及预测研究[D]. 西安: 长安大学, 2014. YI Hui. Study on the evaluation and prediction of expressway asphalt pavement performance [D]. Xian’an: Chan’an University, 2014.
29	骆志元. 基于性能预测的路面养护智能决策研究[D]. 重庆: 西南大学, 2023. LUO Zhiyuan. Research on intelligent decision-making of pavement maintenance based on performance prediction [D]. Chongqing: Southwest University, 2023.
30	GONG H, SUN Y, HU W, et al Investigating impacts of asphalt mixture properties on pavement performance using LTPP data through random forests[J]. Construction and Building Materials, 2019, 204: 203- 212 doi: 10.1016/j.conbuildmat.2019.01.198
31	LI Z, LIU Z, HUANG J, et al MV-GCN: multi-view graph convolutional networks for link prediction[J]. IEEE Access, 2019, 7: 176317- 176328
32	GENG X, LI Y, WANG L, et al. Spatiotemporal multi-graph convolution network for ride-hailing demand forecasting [C]// Proceedings of the AAAI Conference on Artificial Intelligence . Honolulu: AAAI, 2019, 449: 3656–3663.
33	CHEN L, ZHANG H, XIAO J, et al. SCA-CNN: spatial and channel-wise attention in convolutional networks for image captioning [C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition . Honolulu: IEEE, 2017: 6298–6306.
34	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need [C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition . Honolulu: IEEE, 2017: 6000-6010.
35	YAN H, LEI X, PU Z Learning dynamic and hierarchical traffic spatiotemporal features with transformer[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23 (11): 22386- 22399
36	CHUNG J, GULCEHRE C, CHO K, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling [C]// NIPS 2014 Workshop on Deep Learning . Montreal: MIT, 2014.
37	HUBER P J Robust estimation of a location parameter[J]. Breakthroughs in Statistics: Methodology and Distribution, 1992, (2): 492- 518

[1]	刘议丹,朱小飞,尹雅博. 基于异质图卷积神经网络的论点对抽取模型[J]. 浙江大学学报(工学版), 2024, 58(5): 900-907.
[2]	姚鑫骅,于涛,封森文,马梓健,栾丛丛,沈洪垚. 基于图神经网络的零件机加工特征识别方法[J]. 浙江大学学报(工学版), 2024, 58(2): 349-359.
[3]	邓舒文,邵旭东,晏班夫,邱明红. 轻型组合桥梁负弯矩区接缝抗弯性能试验[J]. 浙江大学学报(工学版), 2024, 58(2): 399-412.
[4]	张晗. 用于多元时间序列预测的图神经网络模型[J]. 浙江大学学报(工学版), 2024, 58(12): 2500-2509.
[5]	鲍佳文,高强,唐林,赵唯坚. 钢筋套筒灌浆连接拉伸性能的精细有限元分析[J]. 浙江大学学报(工学版), 2023, 57(4): 814-823.
[6]	周青松,蔡晓东,刘家良. 结合社交影响和长短期偏好的个性化推荐算法[J]. 浙江大学学报(工学版), 2023, 57(3): 495-502.
[7]	曾菊香,王平辉,丁益东,兰林,蔡林熹,管晓宏. 面向节点分类的图神经网络节点嵌入增强模型[J]. 浙江大学学报(工学版), 2023, 57(2): 219-225.
[8]	张京京,张兆功,许鑫. 融合图增强和采样策略的图卷积协同过滤模型[J]. 浙江大学学报(工学版), 2023, 57(2): 243-251.
[9]	周天琪,杨艳,张继杰,殷少伟,郭增强. 基于无负样本损失和自适应增强的图对比学习[J]. 浙江大学学报(工学版), 2023, 57(2): 259-266.
[10]	陆佳炜,李端倪,王策策,徐俊,肖刚. 基于超图卷积神经网络的多行为感知服务推荐方法[J]. 浙江大学学报(工学版), 2023, 57(10): 1977-1986.
[11]	杨燕泽,王萌,刘诚,徐慧通,张小月. 基于语义分割的沥青路面裂缝智能识别[J]. 浙江大学学报(工学版), 2023, 57(10): 2094-2105.
[12]	吕国鹏,蒋楠,周传波,李海波,姚颖康,张旭. 地表爆炸作用下钢筋混凝土管道裂缝扩展机制[J]. 浙江大学学报(工学版), 2022, 56(9): 1704-1713.
[13]	王建锋,邱雨,刘水宙. 基于图像补偿的隧道衬砌裂缝检测方法[J]. 浙江大学学报(工学版), 2022, 56(7): 1404-1415.
[14]	张朝,黄正东,熊仲明,原晓露,许有俊,康佳旺. 地裂缝环境下钢筋混凝土框架结构的地震响应[J]. 浙江大学学报(工学版), 2022, 56(10): 2028-2036.
[15]	任松,朱倩雯,涂歆玥,邓超,王小书. 基于深度学习的公路隧道衬砌病害识别方法[J]. 浙江大学学报(工学版), 2022, 56(1): 92-99.

Viewed

Full text

Abstract

Cited

Shared

Discussed