摘要
针对传统转辙机状态评估方法过度依赖人工经验、参数维度单一等问题,文章以ZDJ9转辙机为研究对象,提出了一种基于改进的转辙机健康状态评估方法。通过构建包含控制电路、启动电路、表示电路、转辙机缺口、环境参数的评价体系,结合大数据深度挖掘技术分析历史故障特征数据,对转辙机的状态性能进行了分析。仿真结果表明,该模型综合评估准确率达92.3%,相较传统的单参数阈值法,故障识别率提升了23%,并实现了机械异常平均提前28 h预警。通过综合分析比较得出转辙机的评估分数,维护人员可根据评估分数对转辙机进行检查。该方法可以减少维护人员因缺乏经验而过度依赖经验员工,提前对隐患转辙机设备进行检查,降低故障发生数量,确保转辙机的正常运行,从而保证列车的安全稳定运营。
To address the limitations of traditional switch machine condition assessment methods,such as excessive reliance on manual experience and single-parameter evaluation,this paper proposes an improved health assessment model using the ZDJ9 switch machine as the research object.An evaluation system was constructed,incorporating control circuits,starting circuits,indication circuits,switch blade gap,and environmental parameters.Historical fault characteristic data were analyzed using big data mining techniques to assess the operational performance of the switch machine.Simulation results demonstrate that the model achieves a comprehensive assessment accuracy of 92.3%,improves the fault identification rate by 23%compared with traditional single-parameter threshold methods,and provides an average early warning of mechanical anomalies 28 hours in advance.By generating an assessment score through comprehensive analysis,maintenance personnel can perform inspections based on the evaluation results.This method reduces the dependence of inexperienced maintenance staff on senior technicians,enables early inspection of potentially faulty equipment,decreases the number of failures,ensures normal operation of switch machines,and enhances the safety and stability of train operations.
作者
陈鹏
CHEN Peng(School of Intelligent Engineering,Chongqing College of Mobile Communications,Chongqing 401520,China)
出处
《智慧轨道交通》
2026年第1期81-86,共6页
INTELLIGENT RAIL TRANSIT
