Search

Localized tooth breakage is one of the common failures of spur gears, which affects the smooth and safe operation of spur gear transmission systems. The tooth collision cannot be neglected, and it is especially important to reveal the meshing-impacting dynamic characteristics of the gear system under localized tooth breakage to improve the safe and stable operation of the gear system. Based on the gear meshing principle and the dissipative collision contact force model, the drive-side tooth meshing model and back-side tooth impacting model are established with considering the transient nature of tooth back-side contact. The tooth surface meshing model and tooth back collision model under partial breakage are established. According to the contact state and force environment of the gear pair, the multi-state meshing-impacting behaviour under local breakage is classified, and the discrete meshing-impact dynamics model of an involute spur gear system under local breakage is established to explore the influence of local breakage on the meshing stiffness and load distribution. The mechanism of contact force under partial tooth breakage is revealed, and the influence of load coefficient and meshing frequency on the nonlinear dynamics is studied by defining two Poincaré maps. It is found that local tooth breakage affects the contact force of single-and double-tooth meshes and reduces gear load carrying capacity. Larger loads inhibit back-side impact, and smaller loads induce the coexistence behaviour and back-side impact. Larger or smaller meshing frequency induce back-side impact behaviour. The coexistence of chaotic and periodic motions induces back-side impact behaviour, and localized tooth breakage affects the coexistence phenomenon and aggravates the complexity of the dynamic behaviour. The dynamic model of gear system considering energy dissipation and nonlinear vibration under the presence of local tooth breakage of the pinion is explored, and the conditions of back-side impact are studied. This research provides new methods and ideas for nonlinear dynamic modelling and analysis of faulty gear systems.

The complexity of urban scenes presents a challenge for semantic segmentation models. Existing models are constrained by factors such as the scale, color, and shape of urban objects, which limit their ability to achieve more accurate segmentation results. To address these limitations, this paper proposes a novel Multi-Scale Feature Shuffle NetWork (MFSNet), which is an improvement upon the existing Deeplabv3+ model. Specifically, MFSNet integrates a novel Pyramid Shuffle Module (PSM) to extract discriminative features and feature correlations, with the objective of improving the accuracy of classifying insignificant objects. Additionally, we propose an efficient feature aggregation module (EFAM) to effectively expand the receptive field and aggregate contextual information, which is integrated as a branch within the network architecture to mitigate the information loss resulting from downsampling operations. Moreover, in order to augment the precision of segmentation boundary delineation and object localization, we employ a progressive upsampling strategy for reinstating spatial information in the feature maps. The experimental results show that the proposed model achieves competitive performance, achieving 80.4% MIoU on the Pascal VOC 2012 dataset, 79.4% MIoU on the Cityscapes dataset, and 40.1% MIoU on the Coco-Stuff dataset. [ABSTRACT FROM AUTHOR]

Object detection methods based on deep learning typically require devices with ample computing capabilities, which limits their deployment in restricted environments such as those with embedded devices. To address this challenge, we propose Mini-YOLOv4, a lightweight real-time object detection network that achieves an excellent trade-off between speed and accuracy. Based on CSPDarknet-Tiny as the backbone network, we enhance the detection performance of the network in three ways. We use a multibranch structure embedded in an attention module for simultaneous spatial and channel attention calibration. We design a group self-attention block with a symmetric structure consisting of a pair of complementary self-attention modules to mine contextual information, thereby ensuring that the detection accuracy is improved without increasing the computational cost. Finally, we introduce a hierarchical feature pyramid network to fully exploit multiscale feature maps and promote the extraction of fine-grained features. The experimental results demonstrate that Mini-YOLOv4 requires only 4.7 M parameters and has a billion floating point operations (BFLOPs) value of 3.1. Compared with YOLOv4-Tiny, our approach achieves a 3.2% improvement in mean accuracy precision (mAP) for the PASCAL VOC dataset and obtains a significant improvement of 3.5% in overall detection accuracy for the MS COCO dataset. In testing with an embedded platform, Mini-YOLOv4 achieves a real-time detection speed of 25.6 FPS on the NVIDIA Jetson Nano, thus meeting the demand for real-time detection in computationally limited devices. [ABSTRACT FROM AUTHOR]

Based on the analysis of notch-root stresses and strains in bodies subjected to multiaxial loading, a quantitative relationship between Neuber rule and the equivalent strain energy density method is found. In the case of elastic range, both Neuber rule and the equivalent strain energy density method get the same estimation of the local stresses and strains. Whereas in the case of elastic-plastic range, Neuber rule generally overestimates the notch-root stresses and strains and the equivalent strain energy density method tends to underestimate the notch-root stresses and strains. A modified method is presented considering the material constants of elastic-plastic Poisson's ratio, elastic modulus, shear elastic modulus, and yield stress. The essence of the modified model is to add a modified coefficient to Neuber rule, which makes the calculated results tend to be more precise and reveals its energy meaning. This approach considers the elastic-plastic properties of the material itself and avoids the blindness of selecting coefficient values. Finally the calculation results using the modified model are validated with the experimental data.

Perturbed eigenvalue equations up to fourth-order are established to detect structural damage in aeroplane wing. Complete set of perturbation terms including orthogonal and non-orthogonal coefficients are computed using perturbed eigenvalue and orthonormal equations. Then the perturbed eigenparameters are optimized using BFGS approach. Finite element model with small to large stepwise damage is used to represent actual aeroplane wing. In small damaged level, termination number is the same for both approaches, while rms errors and termination d-norms are very close. For medium damaged level, termination number is larger for third-order perturbation with lower d-norm and smaller rms error. In large damaged level, termination number is much larger for third-order perturbation with same d-norm and larger rms error. These trends are more significant as the damaged level increases. As the stepwise damage effect increases with damage level, the increase in stepwise effect leads to the increase in model order. Hence, fourth-order perturbation is more accurate to estimate the model solution.