一种改进K-means模型的城市轨道交通突发事件分级方法

针对城市轨道交通事件量化分级的难题，本文提出了一种改进的 K-means聚类的突发事件分级方法.首先，从事件类型、持续时间、影响程度等方面分析各种类型事件的特征规律，提取 8个关键特征量用于聚类分析；其次，应用主成分分析法提取 4个主成分变量并提出权重系数计算方法，实现特征向量降维；提出了基于密度扫描的初始聚类中心确定方法，并将改进的 K-means聚类算法应用于地铁突发事件的分级.案例结果表明，与原始 K-means聚类方法对比，应用本文提出的改进方法聚类效果更佳.研究成果已应用于北京地铁应急指挥系统，验证了本文方法的可行性.

Classification Method of Urban Rail Transit Emergencies Based on Improved K-means Algorithm

This paper presents an improved K-means clustering method for urban rail transit emergencies. Firstly, the characteristics of various types of events are analyzed from the aspects of event type, duration and degree of influence, and 8 key features are extracted for cluster analysis. Secondly, principal component analysis is proposed to extract 4 principal component variables and the weighting coefficient of original variables is calculated. An initial clustering center determination method based on density scanning is proposed, and the improved K-means clustering algorithm is applied to the classification of subway emergency events. Case results show that compared with the original K-means clustering method, the improved method proposed in this paper has better clustering effect. The results were applied in the Beijing subway emergency command system, which verified the feasibility of the method.