| 172 | 0 | 8 |
| 下载次数 | 被引频次 | 阅读次数 |
【背景】快速、准确地识别交通扰动事件的严重程度及其潜在影响，是制定科学应对措施和优化调控策略的前提。然而，现有事件分级评价方法大多依赖于专家知识或人为经验，容易受到主观因素的干扰，限制了评价结果的客观性和有效性。【目标】解决现有交通扰动事件分级评价中主观性较强、自动化水平不足的问题，实现交通扰动事件影响后果的快速、客观分级评价。【方方法法】提出一种基于遗传算法的密度峰值聚类（Genetic Algorithm-based Density Peaks Clustering,GA-DPC）方法。首先，通过识别决策值斜率变化拐点，自动确定初始聚类中心和分类簇数量；其次，构建以最大化轮廓系数Silhouette Index(SI)为目标的优化问题，利用遗传算法求解最优截断距离；最后，基于最优截断距离迭代更新聚类中心和分类簇数量，得到最终聚类结果。【数据】利用公开数据集Spiral数据集、R15数据集和ThreeCircles数据集，以及仿真交通事故和真实降雨扰动事件数据集进行测试。【结果】GA-DPC在公开测试集上的SI值和Calinski-Harabasz(CH)值均优于ADPC、传统DPC、K-means和DBSCAN等聚类算法。在交通事故事件和降雨扰动事件影响的分级评价结果中，GA-DPC同样在SI和CH值上表现出更优的性能，验证了其在不同类型交通扰动事件分级中的有效性。【应用】GA-DPC为交通管理部门提供了一种基于数据驱动的分析工具，能够快速、客观地评估各类扰动事件对交通系统的影响程度，为资源调度和应急管理策略的制定提供决策依据。
Abstract：[Background] The rapid and accurate identification of the severity and potential impact of traffic disturbance events is critical for developing scientific response measures and optimizing management strategies. However, existing event classification methods are heavily reliant on expert knowledge or human experience, making them susceptible to subjective bias, limiting the objectivity and effectiveness of the evaluation results. [Objective] To address the issues of strong subjectivity and insufficient automation in current classification and evaluation methods for traffic disturbance events as well as to achieve rapid and objective grading of the impact of such events. [Methods] A genetic algorithm-based density peak clustering(GA-DPC) method is proposed, which consists of three majorsteps:(1) identifying slope change inflection points in decision values to automatically determine the initial cluster centers and number of clusters,(2) formulating an optimization problem to maximize the Silhouette Index(SI) and solve for the optimal cutoff distance using a genetic algorithm, and(3) iteratively updating the cluster centers and cluster numbers based on the optimal cutoff distance to obtain the final clustering results.[Data] Tests were conducted using public datasets such as Spiral, R15, and ThreeCircles, simulated traffic accident data, and real-world rainfall disturbance data. [Results] On the public test datasets, the GA-DPC method outperformed ADPC, traditional DPC, K-means, and DBSCAN based on the classification metrics SI and Calinski-Harabasz(CH) values. The GA-DPC also demonstrated superior performance in both the SI and CH metrics in the classification and evaluation of traffic accidents and rainfall disturbance events. [Application]The GADPC is a data-driven analytical tool for traffic management departments that allows them to rapidly and objectively assess the severity of traffic disturbance events and their impact on transportation systems. The GA-DPC promotes informed decision-making for resource allocation and the development of emergency management strategies.
[1] HE Y, DU S. Classification of urban emergency based on fuzzy analytic hierarchy process[J]. Procedia engineering,2016, 137:630-638.
[2] REN Q, CHENG G. Application research in accident classification based on fuzzy evaluation system[J]. IOP Conference Series:Materials Science and Engineering, 2019,472(1):012061.
[3]吴凤平,程铁军.基于改进的灰色定权聚类分析的突发事件分级研究[J].中国管理科学, 2013, 21(S1):110-113.WU Fengping, CHENG Tiejun. The classification of emergency based on improved gray clustering model[J].Chinese Journal of Management Science, 2013, 21(S1):110-113.
[4]蔡正杰,梁昌勇,赵树平.突发环境事件等级评估方法研究[J].计算机应用研究, 2014, 31(11):3217-3220.CAI Zhengjie, LIANG Changyong, ZHAO Shuping. Research on emergency environmental accidents grade evaluation[J]. Application Research of Computers, 2014, 31(11):3217-3220.
[5] GONG P, WANG Q, ZHU J. Classification of emergency responses to fatal traffic accidents in Chinese urban areas[J]. Computers, Materials&Continua, 2021, 68(1):1389-1408.
[6]胡思涛,朱艳茹,项乔君.城市出租车行业群发性事件风险预警方法[J].交通运输工程与信息学报, 2014, 12(2):42-48.HU Sitao, ZHU Yanru, XIANG Qiaojun. Forecast of unexpected mass incident risk in urban taxi industry[J]. Journal of Transportation Engineering and Information, 2014,12(2):42-48.
[7]郑宣传,魏运,秦勇,等.一种改进K-means模型的城市轨道交通突发事件分级方法[J].交通运输系统工程与信息, 2019, 19(3):134-140.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.
[8] XING J, ZENG Z, ZIO E. A framework for dynamic risk assessment with condition monitoring data and inspection data[J]. Reliability Engineering&System Safety, 2019,191:106552.
[9]商丽媛,谭清美.基于支持向量机的突发事件分级研究[J].管理工程学报, 2014, 28(1):119-123.SHANG Liyuan, TAN Qingmei. Emergency classification based on support vector machine[J]. Journal of Industrial Engineering and Engineering Management, 2014, 28(1):119-123.
[10]徐绪堪,王京.基于随机森林的突发事件分级模型研究[J].中国安全生产科学技术, 2018, 14(2):77-81.XU Xukan, WANG Jing. Research on classification model of emergencies based on random forest[J]. Journal of Safety Science and Technology, 2018, 14(2):77-81.
[11]刘明辉.基于K-means聚类分析的民航系统恐怖主义风险评估[J].数据分析与知识发现, 2018, 2(10):21-26.LIU Minghui. Risk assessment of civil aviation terrorism based on K-means clustering[J]. Data Analysis and Knowledge Discovery, 2018, 2(10):21-26.
[12] KUMEDA B, ZHANG F, ZHOU F, et al. Classification of road traffic accident data using machine learning algorithms[C]//2019 IEEE 11th International Conference on Communication Software and Networks(ICCSN).Chongqing:IEEE, 2019:682-687.
[13] NAOUALI S, BEN SALEM S, CHTOUROU Z. Clustering categorical data:a survey[J]. International Journal of Information Technology&Decision Making, 2020, 19(1):49-96.
[14] AHMED M, SERAJ R, ISLAM S M S. The k-means algorithm:a comprehensive survey and performance evaluation[J]. Electronics, 2020, 9(8):1295.
[15] PRICE-JONES N, BOVY J. Blind chemical tagging with DBSCAN:prospects for spectroscopic surveys[J].Monthly Notices of the Royal Astronomical Society,2019, 487(1):871-886.
[16] RODRIGUEZ A, LAIO A. Clustering by fast search and find of density peaks[J]. Science, 2014, 344(6191):1492-1496.
[17]陈叶旺,申莲莲,钟才明,等.密度峰值聚类算法综述[J].计算机研究与发展, 2020, 57(2):378-394.CHEN Yewang, SHEN Lianlian, ZHONG Caiming, et al. Survey on density peak clustering algorithm[J]. Journal of Computer Research and Development, 2020, 57(2):378-394.
[18] WANG Y, QIAN J, HASSAN M, et al. Density peak clustering algorithms:a review on the decade 2014—2023[J]. Expert Systems with Applications, 2024, 238:121860.
[19]杨震,王红军,周宇.一种截断距离和聚类中心自适应的聚类算法[J].数据分析与知识发现, 2018, 2(3):39-48.YANG Zhen, WANG Hongjun, ZHOU Yu. A clustering algorithm with adaptive cut-off distance and cluster centers[J]. Data Analysis and Knowledge Discovery, 2018, 2(3):39-48.
[20] WANG Y, WEI Z, YANG J. Feature trend extraction and adaptive density peaks search for intelligent fault diagnosis of machines[J]. IEEE Transactions on Industrial Informatics, 2019, 15(1):105-115.
[21] XU M, LI Y, LI R, et al. EADP:an extended adaptive density peaks clustering for overlapping community detection in social networks[J]. Neurocomputing, 2019,337:287-302.
[22] JING W, JIN T, XIANG D. Fast superpixel-based clustering algorithm for SAR image segmentation[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19:1-5.
[23]许源,陆良,徐冲聪,等.基于改进密度峰值聚类的路网子区动态划分方法[J].交通运输工程与信息学报,2024,22(4):113-127.XU Yuan, LU Liang, XU Chongcong,et al. Dynamic partitioning method for urban road networks based on enhanced density peak clustering[J]. Journal of Transportation Engineering and Information, 2024,22(4):113-127.
[24] SHI L, YANG X, CHANG X, et al. An improved density peaks clustering algorithm based on k nearest neighbors and turning point for evaluating the severity of railway accidents[J]. Reliability Engineering&System Safety,2023, 233:109132.
[25] CHANG H, YEUNG D Y. Robust path-based spectral clustering[J]. Pattern Recognition, 2008, 41(1):191-203.
[26] FR?NTI P, VIRMAJOKI O. Iterative shrinking method for clustering problems[J]. Pattern Recognition, 2006, 39(5):761-775.
[27] DU M, DING S, XU X, et al. Density peaks clustering using geodesic distances[J]. International Journal of Machine Learning and Cybernetics, 2018, 9(8):1335-1349.
基本信息:
DOI：10.19961/j.cnki.1672-4747.2024.12.002
中图分类号:U491
引用信息:
[1]鲍震天,郑凡非,郑芳芳.基于改进密度峰值聚类的交通扰动事件分级评价方法[J].交通运输工程与信息学报,2025,23(02):122-135.DOI:10.19961/j.cnki.1672-4747.2024.12.002.
基金信息:
国家重点研发计划项目（2021YFB1600100）