中文摘要：

为解决牵引供电实时监控中的＂带电/停电/接地＂状态识别和快速计算问题,模拟智能生物的行为特点,提出仿生寻迹算法。该算法基于网络拓扑的图矩阵,推导出拓扑连通的数学模型。在算法实现上,采用标记和追踪技巧,得到快速计算的有关公式。以实际牵引供电系统为例进行数值实验。通过定义计算量因子,比较和分析仿生寻迹算法和幂法的计算量因子和效率。结果表明：仿生寻迹算法比幂法收敛快,计算速度提高约50%。该算法已作为牵引供电运行状态在线监测系统的核心模块投入工程应用。

英文摘要：

In order to identify the power-on/power-off/grounding state and achieve rapid calculation in the traction power supply real-time monitoring system,the bionic tracing algorithm is proposed by simulating intelligent biological properties.The topology connectivity model is established on the basis of the network topology graph matrix.Marking and tracing techniques are applied in algorithm implementation,and related calculation formulas are derived.Numerical experiments are carried out with an actual traction power supply system.By defining the calculation factor,the efficiency of the bionic tracing algorithm and power algorithm is compared and analyzed,and the results have shown that the bionic tracing algorithm converges faster than the power algorithm,and its calculating speed increases by about 50%.The bionic tracing algorithm has been running as a core module of the traction power supply monitoring system to apply to engineer projects.