基于数据挖掘处理的影响辽宁高速公路冬季交通事故主要气象要素分析

doi:10.3969/j.issn.1673-503X.2020.03.008

气象与环境学报 ›› 2020, Vol. 36 ›› Issue (3): 55-63.doi: 10.3969/j.issn.1673-503X.2020.03.008

基于数据挖掘处理的影响辽宁高速公路冬季交通事故主要气象要素分析

林毅¹(),肇毓²,李倩³,马东雷¹,张广梅¹,马强⁴,李岚^1,*(),赵凡¹

1. 辽宁省气象服务中心, 辽宁沈阳 100166
2. 辽宁省高速公路运营管理有限责任公司, 辽宁沈阳 110055
3. 沈阳区域气候中心, 辽宁沈阳 100166
4. 辽宁省交通运输事业发展中心, 辽宁沈阳, 110003

收稿日期:2019-05-17 出版日期:2020-06-30 发布日期:2020-07-09
通讯作者: 李岚 E-mail:linyi_0330@163.com;dy-pisces@163.com
作者简介:林毅,男, 1985年生,高级工程师,主要从事应用气象与气象数据分析。E-mail:linyi_0330@163.com
基金资助:
辽宁省气象局关键技术项目“辽宁省高速公路团雾发生规律及局地性研究”(LNGJ201904);辽宁省气象局科研课题“基于天气影响的辽宁省高速公路交通安全及对策研究”(201613);“基于人工智能的专业气象服务数据挖掘与监控技术研究”(BA201906)

Analysis of main meteorological factors influencing winter traffic accidents in Liaoning expressway based on data mining processing

Yi LIN¹(),Yu ZHAO²,Qian LI³,Dong-lei MA¹,Guang-mei ZHANG¹,Qiang MA⁴,Lan LI^1,*(),Fan ZHAO¹

1. Liaoning Province Public Meteorological Service Center, Shenyang 110166, China
2. Liaoning Province Expressway Operation Administration Co., Ltd., Shenyang 110055, China
3. Shenyang Regional Climate Center, Shenyang 110166, China
4. Liaoning Transportation Development Center, Shenyang 110003, China

Received:2019-05-17 Online:2020-06-30 Published:2020-07-09
Contact: Lan LI E-mail:linyi_0330@163.com;dy-pisces@163.com

摘要/Abstract

摘要：

利用2014—2016年辽宁省冬季高速公路事故记录，将多要素逐日气象观测数据与事故数据按照日期进行匹配，分析气象条件引发高速事故的空间分布特征。数据挖掘分析首先利用两步聚类方法确定辽宁地区冬季天气类型数量，再通过K-means方法对气象数据进行聚类处理。采用随机森林方法对不同天气类型搭建高速公路交通事故分类模型，并分析模型中气象要素的特征重要性。结果表明：受气象条件影响的高速事故数量辽南地区最多，其次是辽西地区，辽东、辽北地区占比较低。辽宁冬季天气可以分为四种类型，根据气象要素数据结构总结出的天气特征分别为：当日出现降水、前一日出现降水、寒冷干燥、潮湿回暖。有明显降水特征天气类型的事故率超过七成，降温、升温天气类型的事故率在两成左右。随机森林方法对前一日出现降水、寒冷干燥两种天气类型的分类精度更高，模型泛化能力也更好。4种天气类型中气象要素的特征重要性有明显差异，最高地温要素在事故高发天气类型中特征重要性排第1位，在潮湿回暖天气类型排第3位，对于冬季辽宁地区高速公路交通安全影响高于其他要素。

关键词: 高速公路, 高风险天气, 数据挖掘, 气象要素

Abstract:

Using the records of winter expressway accidents occurring in Liaoning province and the corresponding multi-factor weather observational data from 2014 to 2016, the spatial distribution characteristics of expressway accidents influenced by meteorological conditions were analyzed.Two-step clustering method was used first in data mining analysis processing to determine the number of winter weather types in Liaoning province, and then the cluster analysis of the meteorological data was performed using K-means method.The random forest method was used to construct an expressway traffic accident classification model for different weather types, and finally, the characteristic importance of meteorological factors in the model was analyzed.The results show that the number of expressway accidents influenced by meteorological conditions is the highest in the southern Liaoning province, followed by the western area, and numbers in the eastern and northern area are relatively low.The winter weather in Liaoning province can be divided into four types.According to the data structure of meteorological factors, the weather characteristics are as follows:precipitation occurring on the day, on the previous day, cold and dry, and warm and humid.The accident rate of weather types with obvious precipitation characteristics exceeds 70%, and the one with cooling and warming weather types is about 20%.The random forest method has high classification accuracy for the two weather types, i.e., precipitation occurring on the day and cold and dry types.The generalization ability of the model is better.The characteristic importance parameters of meteorological factors for the four weather types are significantly different.The surface temperature factor has a greater significance, occupying the first place in the high-incidence weather types, and the third place in the humid warm weather type.

Key words: Expressway, High-risk weather, Data mining, Meteorological factors

中图分类号:

林毅,肇毓,李倩,马东雷,张广梅,马强,李岚,赵凡. 基于数据挖掘处理的影响辽宁高速公路冬季交通事故主要气象要素分析[J]. 气象与环境学报, 2020, 36(3): 55-63.

Yi LIN,Yu ZHAO,Qian LI,Dong-lei MA,Guang-mei ZHANG,Qiang MA,Lan LI,Fan ZHAO. Analysis of main meteorological factors influencing winter traffic accidents in Liaoning expressway based on data mining processing[J]. Journal of Meteorology and Environment, 2020, 36(3): 55-63.

图/表 8

图1

图2

图3

表1

图4

表2

图5

表3

参考文献 24

1	Savolainen P T , Mannering F L , Lord D , et al. The statistical analysis of highway crash-injury severities:a review and assessment of methodological alternatives[J]. Accident Analysis & Prevention, 2011, 43 (5): 1665- 1667.
2	Abegaz T , Berhane Y , Worku A , et al. Effectiveness of an improved road safety policy in Ethiopia:an interrupted time series study[J]. BMC Public Health, 2014, 14 (1): 539. doi: 10.1186/1471-2458-14-539
3	何清, 李宁, 罗文娟, 等. 大数据下的机器学习算法综述[J]. 模式识别与人工智能, 2014, 27 (4): 327- 336. doi: 10.3969/j.issn.1003-6059.2014.04.007
4	Nayak R N, Piyatrapoomi N, Weligamage J.Application of text mining in analysing road crashes for road asset management[C].Proceedings of the 4th World Congress on Engineering Asset Management.London: Springer, 2009: 45-58.
5	Beshah T , Ejigu D , Abraham A , et al. Knowledge discovery from road traffic accident data in ethiopia:data quality, ensembling and trend analysis for improving road safety[J]. Neural Network World, 2012, 22 (3): 215- 244. doi: 10.14311/NNW.2012.22.013
6	Harb R , Yan X D , Radwan E , et al. Exploring precrash maneuvers using classification trees and random forests[J]. Accident Analysis & Prevention, 2009, 41 (1): 98- 107.
7	Nayak R, Emerson D, Weligamage J, et al.Road crash proneness prediction using date mining[C]//Proceedings of the 14th international conference on extending database technology.New York: ACM, 2011.
8	Ho T K.Random decision forests[C]//Proceedings of the 3rd International Conference on Document Analysis and Recognition.Montreal: IEEE, 1995.
9	Breiman L . Random Forests[J]. Machine Learning, 2001, 45 (1): 5- 32.
10	Abdel-Aty M , Pande A , Das A , et al. Assessing safety on dutch freeways with data from infrastructure-based intelligent transportation systems[J]. Transportation Research Record:Journal of the Transportation Research Board, 2008, 2083 (1): 153- 161. doi: 10.3141/2083-18
11	Munde V , Singh D , Singh R , et al. A perspective analysis of traffic accident using data mining techniques[J]. International Journal for Scientific Research & Development, 2014, 2 (3): 1665- 1667.
12	周尔波, 刘澜. 灾害性天气对高速公路交通安全的影响及对策[J]. 交通科技, 2008, (4): 94- 96.
13	林毅, 李倩, 张凯, 等. 气象条件对辽宁省高速公路交通安全的影响研究[J]. 气象与环境学报, 2018, 34 (3): 106- 111. doi: 10.3969/j.issn.1673-503X.2018.03.013
14	张运福, 林毅, 李倩, 等. 气象条件对高速公路交通影响研究进展[J]. 黑龙江交通科技, 2016, 39 (5): 168- 169. doi: 10.3969/j.issn.1008-3383.2016.05.097
15	梁慧玲, 林玉蕊, 杨光, 等. 基于气象因子的随机森林算法在塔河地区林火预测中的应用[J]. 林业科学, 2016, 52 (1): 89- 96. doi: 10.3969/j.issn.1006-1126.2016.01.017
16	Bacher J, Wenzig K, Vogler M.Recent developments and applications in social research methodology[C]//Proceedings of the RC33 Sixth International Conference on Social Science Methodology.Leverkusen: Budrich, 2004.
17	Kov M , Yai T . Exploring Factors Associated with Motorcyclist Accident Involvement in Phnom Penh[J]. Journal of the Eastern Asia Society for Transportation Studies, 2011, 9, 1796- 1811.
18	王伟建, 姚展予, 贾烁, 等. 随机森林算法在人工增雨效果统计检验中的应用研究[J]. 气象与环境科学, 2018, 41 (2): 111- 117.
19	陈喆, 贾春福, 宗楠, 等. 随机森林在程序分支混淆中的应用[J]. 电子学报, 2018, 46 (10): 2458- 2466. doi: 10.3969/j.issn.0372-2112.2018.10.020
20	白琳, 徐永明, 何苗, 等. 基于随机森林算法的近地表气温遥感反演研究[J]. 地球信息科学学报, 2017, 19 (3): 390- 397.
21	Rodriguez-Galiano V F , Ghimire B , Rogan J , et al. An assessment of the effectiveness of a random forest classifier for landcover classification[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2012, 67, 93- 104. doi: 10.1016/j.isprsjprs.2011.11.002
22	邓天宏, 王国安, 焦建丽, 等. 草温、0cm地温、气温间变化规律分析[J]. 气象与环境科学, 2009, 32 (4): 47- 50. doi: 10.3969/j.issn.1673-7148.2009.04.011
23	马晓刚, 胡春丽, 焦敏, 等. 路面湿滑气象等级研究与建立[J]. 气象与环境学报, 2015, 31 (6): 153- 157. doi: 10.3969/j.issn.1673-503X.2015.06.021
24	王春玲, 郭文利, 李迅, 等. 京津冀地区高速公路冰冻灾害风险区划[J]. 气象与环境学报, 2018, 34 (1): 45- 51. doi: 10.3969/j.issn.1673-503X.2018.01.006

气象要素	天气数据集群(n=3920)(列数据为该等级的占比)
气象要素		簇1(n=184)	簇2(n=386)	簇3(n=2200)	簇4(n=1150)
最小能见度/m	< 100	1.6	0.7	1.9	2.2
	100—500	12.0	16.1	3.1	6.7
	> 500	86.4	83.2	95	91.1
平均气温/℃	< -10	0	1.6	16.4	0
	-10—0	55.7	84.2	83.1	28.6
	> 0	44.3	14.2	0.5	71.4
最低气温/℃	< -10	9.2	4.6	54.7	0.3
	-10—0	76	90.2	45.3	68.9
	> 0	14.8	5.2	0	30.8
最高气温/℃	> 0	89.6	87	45.7	98.7
平均相对湿度	> 80	50.3	21.2	6.6	24.8
地面最高温度/℃	> 0	78.1	78	58.9	95.7
地面平均温度/℃	> 0	38.3	22.5	0.1	71.2
地面最低温度/℃	< -5	30.6	18.4	78.7	20.4
	-5~0	56.8	75.9	21.1	62.7
	> 0	12.6	5.7	0.2	16.9
当日降水量/mm	=0	0	83.7	84.4	80.9
	0.1—10	72.7	16.1	15.6	19.1
	> 10	24.3	0.2	0	0
前一日降水量/mm	=0	0	0	70	67.8
	0.1—10	60.1	65	30	32.2
	> 10	39.9	35	0	0
事故率		74.3	88.3	20	22.3

要素	簇1	簇2	簇3	簇4
袋外分数	0.85	0.92	0.95	0.94
模型精度	89.6	98.2	91.3	86.9

簇别	< 0 ℃(事故率)	0—5 ℃(事故率)	> 5 ℃(事故率)
簇1	38(79%)	100(87.1%)	46(43.5%)
簇2	80(80%)	250(95.2%)	56(71.4%)
簇3	862(26.8%)	641(21.7%)	697(7.3%)
簇4	36(38.9%)	226(46%)	888(15.7%)

[1]	王宏,郑秋萍,洪有为,吴建成. 2017年漳州海陆风特征与冬春季污染物浓度关系[J]. 气象与环境学报, 2020, 36(3): 33-40.
[2]	张婉莹,孙秀博,李岚,张建强,孙丽,齐昕. 辽宁沿海高速公路浓雾气候特征及气象影响因子[J]. 气象与环境学报, 2020, 36(1): 67-73.
[3]	郑庆锋, 史军, 董广涛, 黄文娟, 徐卫忠, 王泉源. 上海地区AQI变化特征及与气象因素的相关性[J]. 气象与环境学报, 2019, 35(5): 53-62.
[4]	刘华悦, 付桂琴, 王洁. 石家庄市气象要素对成人哮喘住院人数的影响[J]. 气象与环境学报, 2019, 35(5): 137-143.
[5]	张滢滢, 陈明, 陈丽英. 海口市重污染天气潜在源分析研究[J]. 气象与环境学报, 2019, 35(2): 107-111.
[6]	朱红蕊, 刘赫男, 张洪玲, 尹嫦姣. 哈尔滨市空气质量特征及其与气象要素的关系[J]. 气象与环境学报, 2019, 35(1): 53-58.
[7]	田磊, 裴琳, 陈宇罡, 夏江江, 汪翔. 1970—2015年蚌埠地区霾日气候资料重建及时空变化特征[J]. 气象与环境学报, 2018, 34(5): 39-46.
[8]	胡晓, 徐璐, 蒋飞燕, 俞科爱. 宁波地区不同季节不同强度霾天气变化特征[J]. 气象与环境学报, 2018, 34(5): 31-38.
[9]	赛瀚, 李燕, 刘硕, 全美兰, 孙悦程. 东北区域数值模式产品在天气预报业务中的检验[J]. 气象与环境学报, 2018, 34(5): 119-127.
[10]	王春玲, 郭文利, 李迅, 姜江, 李宏宇, 陈传雷. 京津冀地区高速公路冰冻灾害风险区划[J]. 气象与环境学报, 2018, 34(1): 45-51.
[11]	李杨, 赵胡笳, 马雁军, 朱轶明, 李潇潇. 本溪市大气颗粒物变化特征及其与气象要素的关系[J]. 气象与环境学报, 2017, 33(1): 58-65.
[12]	安娟, 王鹏, 吴振东, 芦晓彤, 卢卓义, 刘春明. 2006-2012年辽阳市PM₁₀时空变化特征[J]. 气象与环境学报, 2016, 32(6): 52-57.
[13]	陈翔翔, 胡磊, 彭王敏子, 刘波. 1960-2014年南昌市霾天气气象要素及主要大气污染物变化特征[J]. 气象与环境学报, 2016, 32(5): 114-121.
[14]	朱轶明, 张晋广, 袁健, 张秀艳, 李洋, 张维全, 万绪江, 卢秉红. 盖州市大气能见度与气象条件相关分析[J]. 气象与环境学报, 2016, 32(4): 55-62.
[15]	顾婷婷, 潘娅英, 骆月珍. 浙江省高速公路低温雨雪冰冻灾害风险评估与区划[J]. 气象与环境学报, 2016, 32(3): 102-106.

基于数据挖掘处理的影响辽宁高速公路冬季交通事故主要气象要素分析

Analysis of main meteorological factors influencing winter traffic accidents in Liaoning expressway based on data mining processing

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 24

相关文章 15

Metrics

本文评价

推荐阅读 7