In order to focus on the injury mechanism of electric cyclists in the collision accident between right turn of truck and electric bicycle, Pc-Crash is used to value the injury of the electric cyclist in this paper. Through repeated tests in Pc-Crash, it was found that two dynamic response process were existed. The electric bicycle and cyclist collided with truck, and the electric bicycle and cyclist were thrown to the ground due to the impact force. Finally, they were crushed under certain conditions. The multi-rigid-body truck model is built on the theories of multi-rigid-body modeling and multi-rigid-body collision to get the value of the crushed part which the single rigid model can’t calculate. After that, we analyze HIC, continuous 3 ms resultant acceleration of chest, the shear force of femur and leg, based on the electric bicycle speed, the mass of electric bicycle and cyclist, the height of cyclist’s seat. We found that when the electric cyclist is not crushed by the truck, the value of each damage index is below the threshold. But when the cyclist is crushed by the truck, the damage index of the part where the cyclist is crushed exceeds the threshold, and causes fatal injury to the cyclist. So, reducing the possibility of crushing can improve the safety level of electric cyclists.

Driver fatigue has a direct impact on urban railway transit (URT) drivers’ driving behavior and can cause driver error. The existing methods for fatigue detection mainly train the models with supervised learning, relying heavily on the annotation of recorded data. However, labeled data are unobtainable in some environments, especially for URT driver fatigue levels during actual driving. Therefore, this study proposes a fatigue detection method using unlabeled heart rate variability data to monitor URT driver fatigue in actual working conditions. By utilizing the existing conclusions with regard to factors contributing to fatigue and physiological changes, this study annotated a small number of samples and then used a novel positive and unlabeled learning algorithm based on nearest neighbors and random forest to divide samples into different fatigue levels. The proposed method was evaluated using the URT driver fatigue data sets collected in the lab. Binary classification achieved an accuracy of 79.0%. However, the accuracy of three-class classification was only 55.7%. In addition, the proposed method performed as well using the field data set as it did using the lab data set. The results show the high generalization performance of the proposed method, which could contribute to addressing the issue of lack of labeled training data for fatigue detection in actual working conditions.

As the passenger flow distribution center cooperating with various modes of transportation, the comprehensive passenger transport hub brings convenience to passengers. With the diversification of passenger travel modes, the passenger flow scale gradually increases, which brings significant challenges to the integrated passenger hub. Therefore, it is urgent to solve the problem of early warning and response to the future passenger flow to avoid congestion accidents. In this paper, the passenger flow GRNN prediction model is proposed, based on the K-means cluster algorithm, and an improved index named BWPs (Between-Within Proportion-Similarity) is proposed to improve the clustering effect of K-means so that the clustering effect of the new index is verified. In addition, the passenger flow data are studied and trained by combining with the GRNN neural network model based on parameter optimization (GA); the passenger flow prediction model is obtained. Finally, the passenger flow of Chengdu East Railway Station has been taken as an example, which is divided into 16 models, and each type of passenger flow is predicted, respectively. Compared with the traditional method, the results show that the model can predict the passenger flow with high accuracy.

This paper proposes an adaptive Perona–Malik filtering algorithm based on the morphological Haar wavelet, which is used for vibration signal denoising in rolling bearing fault diagnosis with strong noise. First, the morphological Haar wavelet operator is utilized to presmooth the noisy signal, and the gradient of the presmooth signal is estimated. Second, considering the uncertainty of gradient at the strong noise point, a strong noise point recognition operator is constructed to adaptively identify the strong noise point. Third, the two‐step gradient average value of the strong noise point in the same direction is used to substitute, and the second derivative is introduced into the diffusion coefficient. Finally, diffusion filtering is performed based on the improved Perona–Malik model. The simulation experiment result indicated that not only the algorithm can denoise effectively, but also the average gradient and second derivative in the same direction can effectively suppress the back diffusion of strong noise points to improve the denoising signal‐to‐noise ratio. The experimental results of rolling bearing show that the algorithm can adaptively filter out strong noise points and keep the information of peak in the signal well, which can improve the accuracy of rolling bearing fault diagnosis.

The proportion of accident between right-turning truck and cyclist straight is smaller than other types of collision accidents. But it is far likely for cyclists to suffer severe injury when it happens. In this paper, we pay attention to the effects of four parameters (truck speed, collision angle, friction coefficient, and truck driver's reaction time) on the dynamic response process and the injury level of cyclist of the accident between right-turning truck and cyclist straight. We found that all the parameters have a great impact on those. Two types of dynamic response processes exist. In step one and step two, the injury values of cyclist were all below their threshold when the truck speed was less than 27 km/h. But when the truck speed is greater than or equal to 27 km/h, the injury value of the chest is higher than 60 g. Meanwhile, HIC and the resultant acceleration of the chest increase with truck speed. Most of the injury value of the crushed part exceeds the fixed value when step three happened. So, slowing down the truck speed and reducing the possibility of cyclist being crushed can improve the safety level of cyclists.