Convolutional Recurrent Neural Network for Fault Diagnosis of High‐Speed Train Bogie

Abstract

Timely detection and efficient recognition of fault are challenging for the bogie of high-speed train (HST), owing to the fact that different types of fault signals have similar characteristics in the same frequency range. Notice that convolutional neural networks (CNNs) are powerful in extracting high-level local features and that recurrent neural networks (RNNs) are capable of learning long-term context dependencies in vibration signals. In this paper, by combining CNN and RNN, a so-called convolutional recurrent neural network (CRNN) is proposed to diagnose various faults of the HST bogie, where the capabilities of CNN and RNN are inherited simultaneously. Within the novel architecture, the proposed CRNN first filters out the features from the original data through convolutional layers. Then, four recurrent layers with simple recurrent cell are used to model the context information in the extracted features. By comparing the performance of the presented CRNN with CNN, RNN, and ensemble learning, experimental results show that CRNN achieves not only the best performance with accuracy of 97.8% but also the least time spent in training model.

Arching Detection Method of Slab Track in High-Speed Railway Based on Track Geometry Data

“…CNN is one of the widely utilized deep learning networks proposed by Lecun et al [54]. It is good at detecting tiny and meaningful spatial features and has the characteristics of sparse weights [55]. In general, CNN is applied to computer vision areas, such as image classification [56] and object detection [57].…”

Method for Fault Diagnosis of Temperature-Related MEMS Inertial Sensors by Combining Hilbert–Huang Transform and Deep Learning

“…In [ 43 ], Yang et al proposed an opening damper evaluation method, which used HHT to convert a 1-D signal into 3-D time-frequency images and used these images for training CNNs in order to evaluate the vibration of power transmission systems. In [ 59 ], Liang et al used the intrinsic mode function (IMF) components of EMD to construct the fault features, which were then used to train the CNN-based FD network, which was called CRNN. In [ 60 ], Chen et al proposed an EMD-based decomposing method, called adaptive sparsest narrow-band decomposition (ASNBD) and applied the ASNBD to FD in roller bearings.…”

Advances in Fault Diagnosis for High-Speed Railway: A Review

“…For time series prediction problems, the long-short-term memory (LSTM) neural network shows superior performance, and a LSTMbased bogie fault prediction method was studied in [25], where the spatial and temporal correlation of fault features can be learned from the original time series signals without any prior knowledge. In [26], a convolutional recurrent neural network (CRNN) was proposed for the HST bogie fault diagnosis, which can simultaneously achieve high accuracy and save time as it inherits the advantages of CNN and Simple Recurrent Unit.…”