摘要
针对当前动车组运行故障动态图像检测系统(TEDS)存在误报率高、人工确认工作量大等问题,提出由采用原始的对比法策略改为分车型、分部件的算法策略,对过车车型及车辆上的部件进行了细分,根据不同的部件及其故障形态进行相应的算法开发。文章给出了实际运用中提高TEDS预报准确性的方案,并进行了可行性分析。
On account of problems such as high false alarm rate,heavy workload of manual confirmation existed in current trouble of moving EMU detection system(TEDS),the algorithm strategy which is used according to vehicle models and components is put forward instead of original contrast strategy,the vehicle models and components on the vehicle are subdivided,the related algorithm development is carried out according to different components and their trouble patterns.In this article,the scheme to improve the accuracy of TEDS forecast in practical application is given,and the feasibility analysis is carried out.
作者
陈刚
CHEN Gang(Locomotive&Rolling Stock Detection Institute of China Railway Shanghai Group Co.,Ltd.,Shanghai 200070,China)
出处
《铁道车辆》
2021年第3期97-100,共4页
Rolling Stock
关键词
TEDS
自动识别
算法策略
TEDS
automatic identification
algorithm strategy
