城市交通突发事件风险致因与后果严重程度判别模型

doi:10.16265/j.cnki.issn1003-3033.2025.02.0676

摘要/Abstract

摘要：

为提升城市交通突发事件后果严重程度判别的准确性,明确突发事件风险致因与后果严重程度的相关关系,构建改进的突发事件后果严重程度判别模型(IDM-ECS)并进行试验验证。首先,基于改进的特征选择算法(IFSA)筛选突发事件风险致因,得到列车兑现率、正点率、日路网客运量等重要风险致因;其次,采用改进的混合受限波尔兹曼机模型(HRBM)计算不同风险致因与后果严重程度的关系,通过比较概率值大小得到风险致因与后果严重程度的判别关系;最后,以轨道交通突发事件数据集作为试验样本进行验证,并从召回率、精确度、F₁值等方面与生成受限波尔兹曼机(GRBM)、随机森林(RF)、深度森林(DF)、轻量梯度提升机(LightGBM)等4个模型进行对比。研究结果表明:列车兑现率、正点率、日路网客运量、5号线断面满载率、10号线断面满载率、信号故障以及车辆故障为7个最优风险致因。IDM-ECS模型平均的召回率为90.55%、精确度为91.89%、F₁值为91.06%,均优于对比模型。

关键词: 城市交通, 突发事件, 风险致因, 后果严重程度, 判别模型, 改进的特征选择算法(IFSA)

Abstract:

In order to improve the accuracy of emergency consequence severity assessment, clarify the correlation between the risk causes and consequence severity in urban traffic emergencies, the improved discrimination model of emergency consequence severity (IDM-ECS) was constructed and experimentally verified. First, based on the IFSA, the risk causes of emergencies were screened to obtain the important risk causes such as train fulfillment rate, punctuality rate, and daily network passenger volume and so on. Secondly, the improved hybrid restricted Boltzmann machine(HRBM) model was used to calculate the relationship between different risk causes and the consequence severity, and the discriminative relationship between risk causes and the consequence severity was obtained by comparing the probability values. Finally, the dataset of rail transit emergencies was used as an experimental sample for validation. The performance was compared with four models, including Generating Restricted Boltzmann Machines (GRBM), Random Forest (RF), Deep Forest (DF), and Light Gradient Boosting Machine (LightGBM), in terms of recall, precision, and F₁ value. The results show that train fulfillment rate, punctuality rate, daily network passenger volume, line 5 section full load rate, line 10 section full load rate, signal failure, and vehicle failure are the seven optimal risk causes. The IDM-ECS model has an average recall of 90.55%, precision of 91.89%, and F₁ value of 91.06%, all of which are better than those of the comparison models.

Key words: urban transit, emergency, risk causes, consequence severity, discrimination model, improved feature selection algorithm (IFSA)

中图分类号:

X928.03事故预防与预测

范博松, 邵春福, 王景升, 刘东. 城市交通突发事件风险致因与后果严重程度判别模型[J]. 中国安全科学学报, 2025, 35(2): 220-226.

FAN Bosong, SHAO Chunfu, WANG Jingsheng, LIU Dong. Model on discriminating risk causes and consequence severity of urban traffic emergencies[J]. China Safety Science Journal, 2025, 35(2): 220-226.

图/表 6

图1

表1

表2

表3

表4

表5

参考文献 16

[1]	YAN Hongyan, GAO Ce, ELZARKA H, et al. Risk assessment for construction of urban rail transit projects[J]. Safety Science, 2019, 118: 583-594. doi: 10.1016/j.ssci.2019.05.042
[2]	WANG Jinghong, YAN Wenyu, XU Han, et al. Investigation of the probability of a safe evacuation to succeed in subway fire emergencies based on Bayesian theory[J]. KSCE Journal of Civil Engineering, 2018, 22(3): 877-886.
[3]	ZHENG Xuanchuan, WEI Yun, QIN Yong, et al. Classification method of urban rail transit emergencies based on improved K-means algorithm[J]. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(3): 134-140.
[4]	SINGH R, HORCHER D, GRAHAM D J, et al. Decomposing journey times on urban metro systems via semiparametric mixed methods[J]. Transportation Research Part C: Emerging Technologies, 2020, 114: 140-163.
[5]	马壮林, 邵逸恒, 舒兰, 等. 多层网络视角下地铁网络脆弱性分析[J]. 中国安全科学学报, 2023, 33(8): 164-172. doi: 10.16265/j.cnki.issn1003-3033.2023.08.0098
	MA Zhuanglin, SHAO Yiheng, SHU Lan, et al. Vulnerability analysis of metro network from perspective of multi-layer network[J]. China Safety Science Journal, 2023, 33(8): 164-172. doi: 10.16265/j.cnki.issn1003-3033.2023.08.0098
[6]	ZHOU Zhihua, JI Feng. Deep forest[J]. National Science Review, 2019, 6(1):74-86. doi: 10.1093/nsr/nwy108
[7]	NICOLA P, LOUISE C, GENSERIK R. Learning about risk: machine learning for risk assessment[J]. Safety Science, 2019, 118: 475-486. doi: 10.1016/j.ssci.2019.06.001
[8]	李志慧, 孙雅倩, 陶鹏飞, 等. 交通事故后的交通运行风险状态等级预测方法[J]. 吉林大学学报:工学版, 2022, 52(1): 127-135.
	LI Zhihui, SUN Yaqian, TAO Pengfei, et al. Prediction method of traffic operation risk level after traffic accident[J]. Journal of Jilin University:Engineering and Technology Edition, 2022, 52(1): 127-135.
[9]	姜万录, 李满, 张培尧, 等. 基于全矢增强深度森林的旋转设备智能故障诊断方法[J]. 中国机械工程, 2022, 33(11): 1324-1335.
	JIANG Wanlu, LI Man, ZHANG Peiyao, et al. Intelligent fault diagnosis method for rotating equipment derived from full vector enhanced deep forest[J]. China Mechanical Engineering, 2022, 33(11): 1324-1335.
[10]	邵怡韦, 陈嘉宇, 林翠颖, 等. 小训练样本下齿轮箱故障诊断:一种基于改进深度森林的方法[J]. 航空学报, 2022, 43(8): 118-132.
	SHAO Yiwei, CHEN Jiayu, LIN Cuiying, et al. Gearbox fault diagnosis under small training samples: an improved deep forest based method[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(8): 118-132.
[11]	范冰倩, 董秉聿, 王彪, 等. 基于深度学习的地铁施工作业人员不安全行为识别与应用[J]. 中国安全科学学报, 2023, 33(1): 41-47. doi: 10.16265/j.cnki.issn1003-3033.2023.01.0874
	FAN Bingqian, DONG Bingyu, WANG Biao, et al. Identification and application of unsafe behaviors of subway construction workers based on deep learning[J]. China Safety Science Journal, 2023, 33(1): 41-47. doi: 10.16265/j.cnki.issn1003-3033.2023.01.0874
[12]	吴海涛, 刘月, 杜彗敏. 小样本条件下地铁运营事故致因推理模型[J]. 中国安全科学学报, 2023, 33(3): 134-140. doi: 10.16265/j.cnki.issn1003-3033.2023.03.1134
	WU Haitao, LIU Yue, DU Huimin. Research on model of subway operation accident's cause under small sample condition[J]. China Safety Science Journal, 2023, 33(3): 134-140. doi: 10.16265/j.cnki.issn1003-3033.2023.03.1134
[13]	YANG Jun, DENG Jiangdong, LI Shujuan, et al. Improved traffic detection with support vector machine based on restricted Boltzmann machine[J]. Soft Computing, 2017, 21(11): 3101-3112.
[14]	LI Chaopeng, WANG Jinlin, YE Xiaozhou. Using a recurrent neural network and restricted Boltzmann machines for malicious traffic detection[J]. NeuroQuantology, 2018, 16(5): 823-831.
[15]	ALDWAIRI T, PERERA D, NOVOTNY M A. An evaluation of the performance of restricted Boltzmann machines as a model for anomaly network intrusion detection[J]. Computer Networks, 2018, 144: 111-119.
[16]	孙兴斌, 孙彦赞, 郑小盈, 等. 面向多类不均衡网络流量的特征选择方法[J]. 计算机应用研究, 2017, 34(2): 568-571,594.
	SUN Xingbin, SUN Yanzan, ZHENG Xiaoying, et al. Feature selection for multi-class imbalanced internet traffic[J]. Application Research of Computers, 2017, 34(2): 568-571,594.

后果严重程度	取值
特别严重(Extremely Serious,ES)	10
严重(Serious,S)	5
较严重(Relatively Serious,RS)	2
不严重(Not Serious,NS)	1

真实分类	判别结果
真实分类	NS	RS	S	ES
NS	a₁₁	a₁₂	a₁₃	a₁₄
RS	a₂₁	a₂₂	a₂₃	a₂₄
S	a₃₁	a₃₂	a₃₃	a₃₄
ES	a₄₁	a₄₂	a₄₃	a₄₄

模型	不严重	较严重	严重	特别严重	均值
DRBM	87.26	82.87	89.00	75.00	83.53
RF	85.02	82.32	87.12	62.50	79.24
DF	90.22	86.70	89.29	87.50	88.43
LightGBM	90.32	89.15	93.05	87.50	90.00
IDM-ECS	91.63	88.70	94.36	87.50	90.55

模型	不严重	较严重	严重	特别严重	均值
DRBM	93.31	73.82	83.33	75.00	81.36
RF	92.85	70.43	83.15	55.56	75.50
DF	95.13	78.70	86.29	70.00	82.53
LightGBM	95.85	79.94	89.80	87.50	88.27
IDM-ECS	95.73	82.16	89.88	100.00	91.89

模型	不严重	较严重	严重	特别严重	均值
DRBM	90.18	78.08	86.07	75.00	82.34
RF	88.76	75.91	85.09	58.82	77.15
DF	92.61	82.51	87.77	77.78	85.16
LightGBM	93.00	84.29	91.40	87.50	89.05
IDM-ECS	93.64	85.30	91.96	93.33	91.06