An equipment multiple failure causes intelligent identification method based on integrated strategy for subway sliding plug door system under variable working condition

Abstract

The subway sliding plug door has been opened and closed frequently for a long time under variable working conditions, and multiple failure are prone to occurring at the same time, resulting in train shutdowns and even major safety accidents. Due to the complex physical mechanism of the door system and the small difference in multi-source monitoring data collected between different states of the same part, it is difficult to identify the multiple failure causes of the door system components by using a single program. Therefore, identifying the multiple failure causes of the key components of the subway sliding plug door has become a challenging problem in the health management of the door system. Aiming at the problem, an equipment multiple failure causes intelligent identification method is proposed based on an integrated strategy by t-distributed stochastic neighbor embedding (TSNE) and particle swarm optimization extreme learning machine (PSO-ELM). In the proposed method, firstly, the sensitive features that can reflect the degradation state of equipment are obtained by using the random forest to measure the importance of features and sort them; Secondly, feature dimensionality is reduced by using TSNE to map the screened high-dimensional features to low-dimensional space. Finally, the parameters of the ELM are optimized by using the PSO algorithm to build a multiple failure causes classification model. The proposed method is verified by the 1:1 benchmark test data of subway sliding plug doors. The results show that the proposed method has higher classification accuracy, faster calculation speed, and stronger generalization ability. The proposed method is an effective integrated strategy to identify multiple failure causes in the subway sliding plug door system and guide the health management and operational maintenance of the subway sliding plug door system.

An enhanced perception health state assessment method for subway sliding plug door transmission system

“…At present, the health state assessment of sliding plug doors mostly uses speed and current signals [19], as shown in figures 4 and 5, which show the rotational speed and current data of the door under normal conditions, respectively, after segmentation according to the control curve. However, the collection of these two-monitoring data is easily affected by complex installation positions, and it is difficult to explain the transmission system fault from the mechanism of the datadriven model, which relies only on the above two monitoring data.…”

Fault detection system of subway sliding plug door based on adaptive EMD method

“…Qi et al [24] presented a health state assessment method of the subway sliding plug door, based on the recognition of rotational speed. An intelligent identification method [25] based on an integrated strategy is proposed for the subway sliding plug door system. When the subway sliding plug door makes a fault, there is a correspondence between the mechanical fault and the motor current.…”

An unsupervised intelligent fault diagnosis research for rotating machinery based on NND-SAM method

“…units, pumps, and other rotating elements [4][5][6][7]. Currently, the early detection of faults in mechanical equipment, particularly fault diagnosis under multiple operating situations, is the most important and challenging aspect of the field of health monitoring; even for the same fault in the same equipment, a change in operating state due to a load of structural elements, pipeline pressure, speed, etc has a severe impact on the accuracy of fault diagnosis in real engineering tasks [8,9]. The majority of researchers have not yet discovered a more effective mechanical fault detection technique for identifying faults in data with larger-scale feature distribution discrepancies.…”