欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为高效评估高速列车运行状态,采用高速列车振动数据时域上的平均值、标准差、有效值等6种统计特征作为特征向量,对列车空气弹簧无气故障状态、列车横向减振器全拆故障状态、列车抗蛇形减振器全拆故障状态、列车正常运行状态进行表征,基于Hadoop平台结合并行化K近邻分类算法进行状态分类评估。实验结果证明,该方法能够很好地评估高速列车的故障状态,有效加快了大数据类问题的分析处理速度。
英文摘要：
To efficiently assess the running status of high-speed trains,the train vibration monitoring data were analyzed and 6 kinds of time-domain statistical characteristics such as mean,standard deviation,RMS were used as the feature vector to achieve the characterization of four kinds of high-speed trains state including train air spring airless fault condition,the train demolished the entire lateral damper failure state,anti-snake train demolished the entire damper fault state,the train normal running state. Platform parallelism K nearest neighbor classification algorithm was used to classify and assess the running status of high-speed trains based on Hadoop.Experimental results show that the method can assess the state of high-speed trains and the paralleliza-tion way can effectively speed up the process of solving big data problem.
同期刊论文项目
同项目期刊论文
期刊信息
位置：