摘要
传统方法在高速动车组故障预警中的应用效果不佳,不仅误报率较高,而且预警时间较长,无法达到预期的预警效果,为此提出基于多尺度信息融合的高速动车组故障分级预警方法。利用温度传感器、电流传感器、转速传感器采集高速动车组故障数据,通过对采集的故障数据进行多尺度信息融合处理,提取动车组故障特征。根据提取的高速动车组故障特征计算动车组故障概率,确定故障预警等级,实现高速动车组故障分级预警。实验结果表明,该方法的误报率在1%以内,平均预警时间为0.38 s,具有良好的可行性与适用性。
The application effect of traditional methods in fault warning of high-speed multiple units is not satisfactory.Not only is the false alarm rate relatively high,but also the warning time is long,which cannot achieve the expected warning effect.Therefore,a multi-scale information fusion based fault classification warning method for high-speed multiple units is proposed.Using temperature sensors,current sensors,and speed sensors to collect fault data of high-speed high-speed trains,multi-scale information fusion processing is performed on the collected fault data to extract fault features of high-speed trains.Calculate the probability of high-speed train unit faults based on the extracted fault features,determine the fault warning level,and achieve high-speed train unit fault classification warning.The experimental results show that the false alarm rate of this method is within 1%,and the average warning time is 0.38 s,indicating good feasibility and applicability.
作者
梁志斌
张永生
LIANG Zhibin;ZHANG Yongsheng(Qingdao Bullet Train Section of China Railway Jinan Bureau Group Co.,Ltd.,Qingdao Shandong 266011,China;CRRC Qingdao Sifang Rolling Stock Co.,Ltd.,Qingdao Shandong 266111,China)
出处
《信息与电脑》
2023年第23期47-49,共3页
Information & Computer
关键词
多尺度信息融合
故障分级预警
高速动车组
multi-scale information fusion
hierarchical fault warning
high-speed EMU
