基于DBSCAN与LightGBM的盾构推进系统故障诊断

Fault Diagnosis of Shield Machine Propulsion System Based on DBSCAN and LightGBM

针对目前现有的盾构推进系统故障诊断算法,大部分无法区分已知故障与未知故障的问题。提出结合DBSCAN与LightGBM的综合盾构推进系统故障诊断算法,通过DBSCAN聚类算法将采样数据分为已知类别数据与未知故障数据,然后将已知类别数据输入LightGBM区分故障类别,两个算法相结合可充分利用工程实际环境中的大量无标签数据与少量的有标签数据,有效区分故障类别。

To solve the problem that most of the existing fault diagnosis algorithms of the shield propulsion system cannot distinguish between known faults and unknown faults,a comprehensive fault diagnosis algorithm combining DBSCAN and LightGBM is proposed.DBSCAN clustering algorithm is applied to divide the sampled data into known data and unknown fault data,and then inputs the known data into LightGBM for fault classification.The combination of this two algorithms can make full use of the large amount of unlabeled data and small amount of labeled data in the actual engineering environment and distinguish fault categories effectively.