期刊文献+

可靠性度量下的列车运行异常自动化预警技术

Automatic Early Warning Technology for Abnormal Train Operation under Reliability Measurement
下载PDF
导出
摘要 对于列车运行异常信号快速变化、且难以确认列车运行失效概率的情况,提出可靠性度量下的列车运行异常自动化预警技术。通过采集列车的运行状态数据,对数据进行去噪处理,可以减少噪声对数据的干扰。去噪处理后,将列车运行数据进行固有模态分解,分解为不同频率范围的子信号。根据子信号不同尺度的变换系数,实现列车状态运行数据重构,还原原始信号。通过对重构后的信号进行分析,获得列车运行状态数据的可靠性度量阈值,用于构建列车运行失效概率近似估计函数。构建估计函数时,通常使用增量系数分析不同列车运行特性和容忍度对可靠性的影响;当失效概率超过自动化可靠性阈值后,自动触发预警流程,实现列车运行异常自动化预警。实验结果表明,该技术应用后,随着列车运行状态数据量的不断增加,运行状态数据中夹杂的噪声数据较少,数据迭代次数增加后,预警时间短且检测结果与实际列车状态一致,可以保证列车运行可靠性。 For situations where abnormal train operation signals rapidly change and it is difficult to confirm the probability of train operation failure,an automated early warning technology for train operation abnormalities under reliability measurement is proposed.By collecting train operation status data and denoising the data,the interference of noise on the data can be reduced.After denoising,the train operation data is decomposed into natural mode decomposition and sub signals of different frequency ranges.Based on the transformation coefficients of different scales of sub signals,the train status operation data is reconstructed to restore the original signal.By analyzing the reconstructed signal,the reliability metric threshold of train operation status data is obtained,which is used to construct an approximate estimation function for train operation failure probability.When constructing estimation functions,incremental coefficients are usually used to analyze the impact of different train operating characteristics and tolerance on reliability;When the failure probability exceeds the automation reliability threshold,the warning process is automatically triggered to achieve automatic warning of abnormal train operation.The experimental results show that with the continuous increase of train operation status data after the application of this technology,there is less noise data mixed in the operation status data.After the number of data iterations increases,the warning time is short and the detection results are consistent with the actual train status,which can ensure the reliability of train operation.
作者 马迪迪 王晨阳 赵洁雪 MA Di-di;WANG Chen-yang;ZHAO Jie-xue(Hefei Urban Rail Transit Corporation,Hefei 230001)
出处 《环境技术》 2023年第11期35-41,共7页 Environmental Technology
关键词 可靠性度量 列车运行异常 自动化预警技术 数据去噪 人工神经网络 reliability measurement abnormal train operation automated early warning technology data denoising artificial neural network
  • 相关文献

参考文献12

二级参考文献125

共引文献16

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部