基于大数据分析的列车轮对故障诊断方法

Research on Fault Diagnosis Method of Train Wheelset Based on Big Data Analysis

随着我国列车行驶速度的不断提高,列车的行车安全逐渐得到了人们的重视,海量列车轮对监测数据为分析列车运行状态提供了条件.为了提高列车轮对故障诊断效率和准确性,文章提出一种基于大数据分析的列车轮对故障诊断方法,针对传统列车轮对故障诊断方法在处理大规模监测数据集时存在处理时间长,故障结果不准确等问题.首先设计一个监测数据融合框架,然后将多故障诊断循环神经网络算法与大数据MapReduce分布式计算框架相结合,利用循环神经网络算法特征提取能力和MapReduce快速计算能力.这样不但能够发挥循环神经网络故障特征提取能力,还能够满足列车轮对故障诊断精确性和实时性的需求,最后通过实例分析,证明了该方法的有效性.

With the improvement of the train speed,more and more importance has been attached to the safety of the train. A large number of train wheelset monitoring data provide conditions to analyze the train running status.To improve the efficiency and accuracy of train wheelset fault diagnosis,based on Big Data analysis,an approach for train wheel sets fault diagnosis was put forward. The aim was to sovle the problems of the traditional train wheelset fault diagnosis method,such as long processing time,and inaccurate fault results. A framework of data fusion for monitoring was presented at begin. Then,combined the multi fault diagnosis using cyclic neural network algorithm with the big data MapReduce distributed computing framework,the ability of feature extraction of the cyclic neural network algorithm and MapReduce fast computing ability was used. Result showed that the method meets the requirement of the accuracy and real-time of the train wheel sets fault diagnosis. The case analysis also proves the effectiveness at later.