特种设备安全事故时间序列拓扑网络特征研究

 为揭示特种设备安全事故时间序列的非线性动力特征，以2005—2020年中国不同类型特种设备安全事故时间序列为对象，采用可视图法将特种设备安全事故时间序列转化为拓扑网络图，生成拓扑网络结构模型；利用拓扑网络理论分析节点度、网络密度、加权聚类系数、幂律指数、介数中心性等拓扑网络特征参数，挖掘特种设备事故发生的时间序列规律。结果表明：各类特种设备安全事故时间序列的拓扑网络具有小世界特性和无标度特性；拓扑网络的聚类系数均较大，社团结构明显；介数中心性越大的节点，对应年份发生事故概率越大。采用的拓扑网络分析方法能更简洁、直观地展示特种设备安全事故时间序列的拓扑网络结构，更全面地表征特种设备安全事故时间序列的非线性动力特征，可为特种设备安全事故预测提供理论基础。

Research on topological characteristics of special equipment safety accidents time series

In order to reveal the nonlinear dynamic characteristics of the time series of special equipment safety accidents, taking the time series of different types of special equipment safety accidents in China from 2005 to 2020 as the research object, the visualization method was used to convert the time series of special equipment safety accidents into a topology network diagram to generate a topology network structural model; using topological network theory to analyze topological network characteristic parameters such as node degree, network density, weighted clustering coefficient, power law index, betweenness centrality, etc., the time series law of special equipment accidents was mined. The results show that: The topological networks of various special equipment safety accident time series have small-world characteristics and scale-free characteristics; the clustering coefficients of the topological networks are all large, and the community structure is obvious; the node with greater betweenness centrality corresponds to the corresponding year, the greater the probability of an accident. The topology network analysis method used in this paper can more concisely and intuitively display the topology network structure of the time series of special equipment safety accidents, and more comprehensively characterize the nonlinear dynamic characteristics of the time series of special equipment safety accidents, which can provide a theoretical basis for the prediction of special equipment safety accidents.