A three-dimensional vehicle-axle box bearing coupling model is established. The model can calculate the contact force in three directions and obtain the dynamic response of axle box bearing under the real vehicle running environment. The load distribution on the double row tapered roller bearing and the vehicle is analyzed, and the co-simulation is conducted by comprehensively considering the force transmission between vehicle and bearing. Taking into account the great impact of defects on the bearing, three different types of bearing defects are added into the model, respectively. The simulation results verify the effectiveness of the model. The model is also verified by using the rolling and vibrating test rig of single wheelset. It is concluded that the simulation results show good agreement with experimental results. The influence of track irregularity on the system motion state is studied by using axis trajectory and vibration RMS (Root Mean Square value). The results show that the influence of track irregularity and wheel flat scar on axle box bearing cannot be ignored. The RMS of acceleration will change greatly due to the existence of defects. Wheel flat scar will greatly interfere with the extraction of bearing defect, but it can be selected at high speed and low frequency to monitor the existence of wheel flat scar, and select low speed and high frequency to monitor the existence of bearing defect. The research results are helpful to the detection of wheel flat scar and axle box bearing defect.

Using the magnetorheological (MR) damper model, this paper derives a semiactive suspension model for a high‐speed railway vehicle, and a new evaluating method is proposed to analyze the effect of two kinds of time delay existing in control systems on vehicle dynamic performance. The railway vehicle is modeled by a 50 degree‐of‐freedom (DOF) system which considers the full 6 DOF of each wheelset, bogie, car body, and the pitch angle of each axle box. Several control strategies, sky‐hook (SH), acceleration‐driven damping (ADD), and mixed SH‐ADD, are considered in the semiactive suspension system. To evaluate the effect of these semiactive controls and the different kinds of time delay on the lateral ride index of a high‐speed railway vehicle, a 3D surface in a rectangular coordinate system is described. The cross curve between the 3D surface and a horizontal plane which represents the performance of passive suspension is projected on the X‐Y plane, and the area enclosed by the contour line, X‐axis, and Y‐axis can be used to evaluate the performance of semiactive controls. The results show that the new method is convenient to evaluate the performance of semiactive control strategies visually when there is more than one kind of time delay.

As a sustainable mode of metro-development strategy, transit-oriented development (TOD) is rapidly growing to finance the transport infrastructure investment. The main negative consequence of constructing residential buildings directly over metro depots is railway-induced vibration, that may affect structural serviceability. The residents may feel uncomfortable, as the metro trains start running very early in the morning and finish daily operations very late at night. In order to evaluate the level of human comfort subject to the special situation, a case study was provided in this paper. Directed by the academic review, there were four common comfort evaluation methods, with difference indexes to describe the influence of vibrations. Therefore, a measurement campaign was conducted and both acceleration and velocity sensors were simultaneously installed at the same measurement points, to reduce the influence of the conversion accuracy. The results show that there are certain differences between the evaluation methods in assessing the vibration comfort, but considering the most adverse effects together, the over-track building at this particular TOD-developed depot can ensure that 90% of the occupants would not be highly annoyed by the vibrations. The main negative effect on human comfort at the TOD depot is that the high-level vibrations would cause interruptions in sleep. Among them, the vibrations in this case would affect the rest of 17% of the occupants in the bedrooms on the seventh floor, and make it difficult for 9% of the occupants to fall asleep. Therefore, the evaluation index was suggested to consider more factors related to sleep difficulties and awake threshold values.

The degradation process of bearing performance in the whole life cycle is a complex and nonlinear process. However, the traditional neural network-based approaches usually consider the degradation process of bearing performance as linear, which does not accord with the actual situation of bearing degradation. To overcome this shortcoming, a rolling bearing’s remaining useful life prediction method based on a Squeeze-and-Excitation-Convolutional long short-term memory (SE-ConvLSTM) neural network was proposed based on the construction of a new health index in the process of bearing life evolution. The proposed method considered the change rule of the health indicator during the whole life cycle evolution of bearings, then constructed the health indicator by using the SE-ConvLSTM neural network, effectively improving the model prediction accuracy and training efficiency. Firstly, the original data are filtered and denoised by Ensemble Empirical Mode Decomposition. Combined with Principal Component Analysis (PCA) dimensionality reduction and the Local Outlier Factor (LOF) algorithm, the bearing’s life evolution interval is divided. Then, the health indicator is constructed based on the proposed SE-ConvLSTM model, and the remaining useful life of rolling bearings is predicted by a particle filter and double exponential model. The proposed method is compared with other related methods with the PHM2012 dataset, and the results show that the proposed method has higher accuracy in remaining useful life predictions. Compared with the traditional method, the health index construction based on the division of the lifespan evolution interval has higher practical significance.

Much has been written on the nature of labour flexibility in the Western context and the extent to which it benefits employers in terms securing them cost-effective operations and flexible workers by offering them contingent work arrangements. Absent in this debate has been any examination of the nature and extent of labour flexibility in the non-Western context. This article aims to broaden the debate and examines the current application of labour flexibility practices and its resultant implications in the novel context of Taiwan – with a particular focus on the hospitality industry. The choice of hospitality industry is in line with the recent CEPD's¹ call for labour dispatch agencies to be considered as a promotional service industry among 12 categories of services. Data derived from focus group studies and individual in-depth interviews at four hotels and their partner labour dispatch agencies elicited the triangular relationship among labour dispatch agencies, client hotels, and agency workers. In contrast to previous similar research of the Western context where labour flexibility was primarily seen to secure lower labour costs, it was found that tight managerial control over the flexible workforce plays a crucial role in adopting contingent work arrangements. Moreover, the results indicate that flexible workers are regarded as a cost rather than being considered as the rhetoric of human capital.

Aiming at the problem of insufficient bearing fault samples in practical engineering, an unsupervised bearing fault diagnosis method with double frame twin network is proposed in this paper. This method only uses the source domain data to diagnose the target domain data, and can minimize the problem of domain difference. The concrete method is firstly to carry on continuous wavelet transform to the rolling bearing signal, and the processed signal is used as the input feature of convolutional neural network. The twin network structure is optimized and Manhattan distance is used as the loss function in the training process. The proposed CM network is applied to the diagnosis of inner ring fault, outer ring fault and normal state of rolling bearing. The results show that CM network can achieve over 90% accuracy on SQ variable speed vibration signal data set (SQV), Shijiazhuang Railway University data set (STD) and Case Western Reserve University data set (CWRU), which is suitable for bearing fault diagnosis under different working conditions and provides a new direction for fault diagnosis. The PyTorch code is available at https://github.com/xiaotianqu/The-unsupervised-bearing-fault-diagnosis-method-based-on-the-dual-framework-Siamese-network.