MAB-DrNet: Bearing Fault Diagnosis Method Based on an Improved Dilated Convolutional Neural Network.

Rolling bearing fault diagnosis is of great significance to the safe and reliable operation of manufacturing equipment. In the actual complex environment, the collected bearing signals usually contain a large amount of noises from the resonances of the environment and other components, resulting in the nonlinear characteristics of the collected data. Existing deep-learning-based solutions for bearing fault diagnosis perform poorly in classification performance under noises. To address the above problems, this paper proposes an improved dilated-convolutional-neural network-based bearing fault diagnosis method in noisy environments named MAB-DrNet. First, a basic model called the dilated residual network (DrNet) was designed based on the residual block to enlarge the model's perceptual field to better capture the features from bearing fault signals. Then, a max-average block (MAB) module was designed to improve the feature extraction capability of the model. In addition, the global residual block (GRB) module was introduced into MAB-DrNet to further improve the performance of the proposed model, enabling the model to better handle the global information of the input data and improve the classification accuracy of the model in noisy environments. Finally, the proposed method was tested on the CWRU dataset, and the results showed that the proposed method had good noise immunity; the accuracy was 95.57% when adding Gaussian white noises with a signal-to-noise ratio of -6 dB. The proposed method was also compared with existing advanced methods to further prove its high accuracy.

Heuristic Routing Algorithms for Time-Sensitive Networks in Smart Factories.

Over recent years, traditional manufacturing factories have been accelerating their transformation and upgrade toward smart factories, which are an important concept within Industry 4.0. As a key communication technology in the industrial internet architecture, time-sensitive networks (TSNs) can break through communication barriers between subsystems within smart factories and form a common network for various network flows. Traditional routing algorithms are not applicable for this novel type of network, as they cause unnecessary congestion and latency. Therefore, this study examined the classification of TSN flows in smart factories, converted the routing problem into two graphical problems, and proposed two heuristic optimization algorithms, namely GATTRP and AACO, to find the optimal solution. The experiments showed that the algorithms proposed in this paper could provide a more reasonable routing arrangement for various TSN flows with different time sensitivities. The algorithms could effectively reduce the overall delay by up to 74% and 41%, respectively, with promising operating performances.

A Multi-Objective Task Scheduling Strategy for Intelligent Production Line Based on Cloud-Fog Computing.

With the widespread use of industrial Internet technology in intelligent production lines, the number of task requests generated by smart terminals is growing exponentially. Achieving rapid response to these massive tasks becomes crucial. In this paper we focus on the multi-objective task scheduling problem of intelligent production lines and propose a task scheduling strategy based on task priority. First, we set up a cloud-fog computing architecture for intelligent production lines and built the multi-objective function for task scheduling, which minimizes the service delay and energy consumption of the tasks. In addition, the improved hybrid monarch butterfly optimization and improved ant colony optimization algorithm (HMA) are used to search for the optimal task scheduling scheme. Finally, HMA is evaluated by rigorous simulation experiments, showing that HMA outperformed other algorithms in terms of task completion rate. When the number of nodes exceeds 10, the completion rate of all tasks is greater than 90%, which well meets the real-time requirements of the corresponding tasks in the intelligent production lines. In addition, the algorithm outperforms other algorithms in terms of maximum completion rate and power consumption.

Preliminary evaluation of sagittal and transverse plane cross-sectional variations of the trunk during quiet and deep breathing by optical reflective motion analysis in patients with scoliosis.

Trunk movements during quiet and deep breathing in untreated and in operated patients with scoliosis are not well defined. To evaluate sagittal and transverse plane cross-sectional variations of the trunk during quiet and deep breathing by optical reflective motion analysis (ORMA) in children with scoliosis. Twenty-one patients were divided into three groups: normal subjects (A; n = 6), subjects with untreated scoliosis >50° (B; n = 7) and operated patients (C; n = 8). Standing and sitting height, T1-T12 and L1-L5 length, arm span, chest perimeter, weight and BMI were recorded. Trunk movements of all patients, during quiet and deep breathing, were measured with a 10-camera 3D ORMA system (82 markers) with the subjects in a standard standing position. Groups were comparable with respect to age, sex, height, arm span and weight (P > 0.05). Significant differences were found in Cobb angle, chest perimeter and BMI (P < 0.05). Trunk sagittal and transverse plane movements during quiet and deep breathing decreased significantly in group B and group C when compared to group A (P < 0.05). Surgery does not completely eliminate sagittal and transverse plane kinematics, although the spine is rendered more rigid. This preliminary study involving a relatively limited number of patients outlines ORMA is a useful tool for analyzing sagittal and transverse plane motion abnormalities of the trunk. Trunk kinematics was altered during breathing in unoperated patients with scoliosis >50°. Operated subjects had sagittal plane trunk kinematics closer to normal subjects, although changes were less pronounced at the level of the convex side.

Effects of posterior hemivertebra resection and short segment fusion on the evolution of sagittal balance in children with congenital scoliosis.

There is a paucity of data describing sagittal alignment changes in children with congenital scoliosis (CS) treated by hemivertebra (HV) resection. This study aimed to evaluate the effects of posterior HV resection on spine sagittal alignment in children with CS. This is a retrospective analysis of 31 children with CS (mean age at surgery: 49.61 ± 10.21 months; range, 39-72; mean follow-up time: 5.16 ± 1.21 years; range: 3-7) treated at our Institution. Only patients with single thoracic or single lumbar, fully segmented HV managed by posterior HV resection and two segments fusion with four screws and two robs were included. According to the anatomical location of the HV, patients were divided into two groups: thoracic (group A) and lumbar (group B). Thoracic kyphosis (T1-T12; TK) and lumbar lordosis (L1-S1; LL) were measured pre- and postoperatively at 6 months interval. Postoperative TK and LL were 30.3 ± 11.47 and 28.8 ± 9.47, and were 31.98 ± 9.66 and 46.7 ± 11.37 at the last follow-up visit, respectively. The incidence of thoracic hypokyphosis in group B was 53.3%, and it was significantly higher compared to group A (12.5%, P = 0.04). During follow-up, TK changes were comparable between the two groups of patients while LL improved in all patients 6 months after surgery, and continued to improve thereafter. Posterior HV resection and short segment fusion have limited impact on the evolution of TK; in particular, children with lumbar HV were more likely to be hypokyphotic preoperatively, but less likely postoperatively with an increase in LL and a stabilization of TK.

Joint Soft-Hard Attention for Self-Supervised Monocular Depth Estimation.

In recent years, self-supervised monocular depth estimation has gained popularity among researchers because it uses only a single camera at a much lower cost than the direct use of laser sensors to acquire depth. Although monocular self-supervised methods can obtain dense depths, the estimation accuracy needs to be further improved for better applications in scenarios such as autonomous driving and robot perception. In this paper, we innovatively combine soft attention and hard attention with two new ideas to improve self-supervised monocular depth estimation: (1) a soft attention module and (2) a hard attention strategy. We integrate the soft attention module in the model architecture to enhance feature extraction in both spatial and channel dimensions, adding only a small number of parameters. Unlike traditional fusion approaches, we use the hard attention strategy to enhance the fusion of generated multi-scale depth predictions. Further experiments demonstrate that our method can achieve the best self-supervised performance both on the standard KITTI benchmark and the Make3D dataset.

[Parathyroid hormone inhibits the apoptosis of osteoblast MC-3T3E1 cells through a non-PLC-dependent protein kinase C pathway].

OBJECTIVE: To investigate the effect of the non-PLC-dependent protein kinase C (PKC) pathway of parathyroid hormone (PTH) on the apoptosis and proliferation of osteoblast MC-3T3E1 cells. METHODS: MC-3T3E1 cells were seeded in 96-well plates at the density of 1.5×10(4) cells/mL and incubated for 3 day. The cells were then exposed to 100 nmol/L of [Gly(1), Arg(19)]hPTH(1-28), 100 nmol/L of [Gly(1), Arg(19)]hPTH(1-34), 100 nmol/L of [Gly(1), Arg(19)]hPTH(1-34)+1 µmol/L Go6983, 1 µmol/L Go6983, or deionized water (control) for 1, 24 or 48 h. After the treatments, cell counting kit-8 (CCK-8) and Caspase-Glo® 3/7 Assay (Caspase-3) were used to examine the proliferation and apoptosis of MC3T3-E1 cells. RESULTS: CCK-8 results showed that hPTH(1-34) increased the number of MC3T3-E1 cells compared with hPTH(1-34)+Go6983 at 1 h and 24 h, but this difference was not statistically different. At 48 h, treatment with hPTH(1-34), as compared with hPTH(1-28), significantly increased the number of MC3T3-E1 cells (P<0.05), and this effect was blocked by the PKC inhibitor Go6983 (P<0.05). hPTH(1-34) did not result in significant inhibition of MC3T3-E1 cell apoptosis at 1 h and 24 h as compared with hPTH(1-34)+Go6983, but significantly inhibited the cell apoptosis as compared with hPTH(1-28) (P<0.05); this inhibitory effect was blocked by Go6983 (P<0.05). CONCLUSION: s A relatively long time (for 48 h) of exposure to PTH can inhibit apoptosis and promote the proliferation of MC3T3-E1cells through a non-PLC-dependent PKC pathway.