Node importance recognition algorithm for urban rail transit networks based on Kolmogorov entropy weight
In recent years,the frequency of accidents,such as delays,sabotage events,and operational failures,in China's urban rail transit system has escalated.Most of these incidents occur at locations with high passenger flow and heavy traffic,emphasizing the critical need to identify important nodes in urban rail transit networks for ensuring their normal operation.To realistically model the real-world complexities of such networks,this study initially develops a model based on complex net-work theory.Subsequently,we incorporate the influence of station-specific attributes and surround-ing regional environment on node importance,constructing an index evaluation system for urban rail transit networks.Then,leveraging the commonalities between chaotic time series and chaotic system index series,we introduce the Kolmogorov entropy weight method to compute index weights.The importance ranking of urban rail transit network nodes is then derived by integrating the index data.Case-study results from the Chengdu metro network demonstrate that the top-10 key stations are con-centrated in the central urban area and serve as transfer stations.Comparative analysis with the infor-mation entropy weight method reveals that the proposed approach shares eight common key sites,displaying a more concentrated distribution of ranking results with fewer extreme values and a small-er score gap.These research findings enhance our understanding of the structural characteristics of urban rail transit networks,providing a crucial theoretical foundation for accident prevention and op-erational optimization.
国家科技期刊平台
NSTL
万方数据
