Search

Early fault diagnosis of rotating machinery is crucial in the industry. The network parameters of the traditional deep learning-based fault diagnosis method are optimized only by the single loss function, and the extracted features are usually not the most valuable features of the input signal. This article proposes a novel method, attention recurrent autoencoder (AE) hybrid model classification algorithm, for early fault diagnosis and severity detection of rotating machinery. The AE can learn the most valuable features in an unsupervised way. By adjusting the weight proportion of the two loss functions and optimizing the multibranch network simultaneously, the proposed method enables the network to extract the most valuable features of the input signals. Moreover, the proposed method can take the raw 1-D vibration signal as the input and does not need time–frequency conversion. By introducing long short-term memory networks in AE, the time-dependent features of the data can be extracted effectively. The proposed method was verified by five kinds of gears at different pitting degrees under six load conditions. The results indicate that the method can provide simultaneous accurate fault diagnosis and severity detection for different pitting degrees.

Pitting corrosion of rotating machinery is one of the most common faults in industrial engineering. The convolutional neural network (CNN) is increasingly applied to the fault diagnosis. However, the conventional CNN method will reduce the feature dimension of the collected signal and cause the loss of information during the pooling process. In this paper, a new method based on dilated CNN combined with spatial dropout (DCSD) is proposed to diagnose the early faults of rotating machinery. By filling the convolution kernel, the DCSD method can increase the receptive field of the CNN without increasing the number of parameters while retaining more features of the raw vibration signal of the rotating machine. To avoid the dropout method eliminates the adjacent elements with a strong correlation, the Spatial Dropout method is adopted to reduce the overfitting problem of deep networks. The early pitting gears experiment was designed to verify the DCSD method in this paper. The raw vibration signal data of 6 different healthy states were collected to verify the effectiveness of the method. The experimental results show that the DCSD method proposed can effectively distinguish the different early gears pitting, and the diagnostic accuracy is better than other popular deep learning methods.

Background: Concentric and eccentric cardiac hypertrophy are associated with pressure and volume overload, respectively, in cardiovascular disease both conferring an increased risk of heart failure. These contrasting forms of hypertrophy are characterized by asymmetrical growth of the cardiac myocyte in mainly width or length, respectively. The molecular mechanisms determining myocyte preferential growth in width versus length remain poorly understood. Identification of the mechanisms governing asymmetrical myocyte growth could provide new therapeutic targets for the prevention or treatment of heart failure. Methods: Primary adult rat ventricular myocytes, adeno-associated virus (AAV)–mediated gene delivery in mice, and human tissue samples were used to define a regulatory pathway controlling pathological myocyte hypertrophy. Chromatin immunoprecipitation assays with sequencing and precision nuclear run-on sequencing were used to define a transcriptional mechanism. Results: We report that asymmetrical cardiac myocyte hypertrophy is modulated by SRF (serum response factor) phosphorylation, constituting an epigenomic switch balancing the growth in width versus length of adult ventricular myocytes in vitro and in vivo. SRF Ser 103 phosphorylation is bidirectionally regulated by RSK3 (p90 ribosomal S6 kinase type 3) and PP2A (protein phosphatase 2A) at signalosomes organized by the scaffold protein mAKAPβ (muscle A-kinase anchoring protein β), such that increased SRF phosphorylation activates AP-1 (activator protein-1)-dependent enhancers that direct myocyte growth in width. AAV are used to express in vivo mAKAPβ-derived RSK3 and PP2A anchoring disruptor peptides that block the association of the enzymes with the mAKAPβ scaffold. Inhibition of RSK3 signaling prevents concentric cardiac remodeling induced by pressure overload, while inhibition of PP2A signaling prevents eccentric cardiac remodeling induced by myocardial infarction, in each case improving cardiac function. SRF Ser 103 phosphorylation is significantly decreased in dilated human hearts, supporting the notion that modulation of the mAKAPβ-SRF signalosome could be a new therapeutic approach for human heart failure. Conclusions: We have identified a new molecular switch, namely mAKAPβ signalosome–regulated SRF phosphorylation, that controls a transcriptional program responsible for modulating changes in cardiac myocyte morphology that occur secondary to pathological stressors. Complementary AAV-based gene therapies constitute rationally-designed strategies for a new translational modality for heart failure.

This study was aimed to investigate the prevalence and factors associated with anxiety and depressive symptoms among hospitalized patients with COVID-19 during the epidemic outbreak in Wuhan, China. A total of 99 COVID-19 patients were recruited and completed the Hospital Anxiety and Depression Scale (HADS) and the modified Medical Research Council (mMRC) Scale. Results showed there was no significant difference in anxiety or depressive symptoms between male and female. Patients aged 46–60 years old had a higher ratio of both anxiety and depressive symptoms. Besides, patients whose hospital stays was longer than 14 days had a higher risk of depressive symptoms than those stays was less than 7 days. There was no significant difference in the correlation between level of dyspnea and the levels of anxiety or depressive symptoms. In conclusion, COVID-19 patients might have anxiety and depressive symptoms during hospitalization. Clinicians should pay attention to the middle age group and patients with longer hospital stays.

In recent years, deep learning based diagnostic approaches have become more attractive. However, most of these methods are supervised diagnostic approaches. Developing a supervised diagnostic model requires a large number of labeled training data. And it is time consuming and labor intensive to obtain labeled data for a variety of systems and working conditions. Therefore, an unsupervised diagnostic model that does not require labeled training data is more desirable. This paper proposes an unsupervised diagnostic model by integrating a sparse autoencoder, a deep belief network, and a binary processor. In comparison with the existing unsupervised methods, the proposed method does not need to perform statistical features extraction, and directly uses the normalized frequency domain signals as the inputs. Moreover, in the proposed diagnostic model, the input data is passed through layer by layer without fine-tuning, which is completely unsupervised process. The proposed methods have been validated with bearing fault datasets and gear pitting fault datasets. The validation results show that the proposed method has a higher accuracy for both bearing and gear pitting fault diagnosis.

In aerospace industry, gears are the most common parts of a mechanical transmission system. Gear pitting faults could cause the transmission system to crash and give rise to safety disaster. It is always a challenging problem to diagnose the gear pitting condition directly through the raw signal of vibration. In this paper, a novel method named augmented deep sparse autoencoder (ADSAE) is proposed. The method can be used to diagnose the gear pitting fault with relatively few raw vibration signal data. This method is mainly based on the theory of pitting fault diagnosis and creatively combines with both data augmentation ideology and the deep sparse autoencoder algorithm for the fault diagnosis of gear wear. The effectiveness of the proposed method is validated by experiments of six types of gear pitting conditions. The results show that the ADSAE method can effectively increase the network generalization ability and robustness with very high accuracy. This method can effectively diagnose different gear pitting conditions and show the obvious trend according to the severity of gear wear faults. The results obtained by the ADSAE method proposed in this paper are compared with those obtained by other common deep learning methods. This paper provides an important insight into the field of gear fault diagnosis based on deep learning and has a potential practical application value. Keywords: Deep learning, Gear pitting diagnosis, Gear teeth, Raw vibration signal, Semi-supervised learning, Sparse autoencoder

Fibroblast growth factor (FGF) 21, a hormone that increases insulin sensitivity, has shown promise as a therapeutic agent to improve metabolic dysregulation. Here we report that FGF21 directly targets cardiac myocytes by binding β-klotho and FGF receptor (FGFR) 4. In combination with high glucose, FGF21 induces cardiac myocyte growth in width mediated by extracellular signal-regulated kinase 1/2 (ERK1/2) signaling. While short-term FGF21 elevation can be cardio-protective, we find that in type 2 diabetes (T2D) in mice, where serum FGF21 levels are elevated, FGFR4 activation induces concentric cardiac hypertrophy. As T2D patients are at risk for heart failure with preserved ejection fraction (HFpEF), we propose that induction of concentric hypertrophy by elevated FGF21-FGFR4 signaling constitutes a novel mechanism promoting T2D-associated HFpEF and that FGFR4 blockade might serve as a cardio-protective therapy in T2D. In addition, potential adverse cardiac effects of FGF21 mimetics currently in clinical trials should be investigated.

As the total mileage of the pipeline network continues to increase, the topological structure of the natural gas pipeline network will become more and more complex, and then gradually show a trend of system complexity. The complicated topological structure of the pipeline network is likely to cause inherent structural defects in the pipeline network, which has an impact on the safe operation of the pipeline network. At present, research aimed at ensuring the safe and reliable operation of natural gas pipeline networks has become a research hotspot, but very few studies have considered the complexity of natural gas pipeline networks and its possible impacts. In order to understand the complexity of natural gas pipeline network and its behavior when facing structural changes, this paper studies the robustness of natural gas pipeline network based on complex network theory. This paper draws on the methods and experience of complex network robustness research in other research fields, combined with the operation characteristics of natural gas pipeline network, and proposes a robustness evaluation method suitable for natural gas pipeline network. The robustness evaluation method of natural gas pipeline network is helpful to identify the key components of the pipeline network and understand the response of the pipeline network to structural changes, so as to provide a theoretical reference for the safe and stable operation of the natural gas pipeline network. The evaluation results show that the pipeline network can show strong robustness when faced with random attacks represented by pipe accidents or component failures caused by natural disasters, and when faced with targeted attacks represented by terrorist attacks, the robustness of the natural gas pipeline network is very weak. The pipeline network behaves differently in the face of different types of random attacks. The pipeline network is more robust when faced with random attacks represented by component failures than when faced with random attacks represented by pipe accidents caused by natural disasters.

This study aimed to assess the association of coagulation-related indicators such as plasma fibrinogen (FIB), d-dimer, and fibrin degradation product (FDP) in rheumatoid arthritis (RA) with the disease activity. Data from 105 RA patients and 102 age- and gender-matched healthy controls were collected in the retrospective study. Disease activity score in 28 joints based on C-reactive protein (DAS28-CRP) was used to divide RA patients into low activity group (DAS28-CRP ≤ 2.7) and active group (DAS28-CRP > 2.7). Receiver operating characteristic (ROC) curve was applied to determine area under the curve (AUC). The association between plasma FIB, d-dimer, and FDP and DAS28-CRP was evaluated by spearman correlation. Logistical regression analysis was used to identify the independent variables associated with RA disease activity. RA patients showed higher levels of plasma FIB, d-dimer, and FDP than the controls (P d-dimer, and FDP were also increased in active groups of RA patients than those in inactive groups (P d-dimer was higher than erythrocyte sedimentation rate (ESR) and rheumatoid factor (RF), and that of FDP was higher than RF in RA patients. In addition, the optimal cut-off value of plasma FIB, d-dimer, and FDP for RA diagnosis was 286 mg/dL, 470 μg/L, and 1.45 mg/L, respectively. Spearman analysis showed that plasma FIB, d-dimer, and FDP were positively related with DAS28-CRP (P d-dimer (odds ratio 2.862, 95% confidence interval 1.851–5.426, P d-dimer, and FDP were increased in RA patients and positively correlated with the disease activity of RA. d-dimer may act as a novel inflammatory indice for indicating disease activity in RA patients.

Fault diagnosis of rotating machinery is essential in the modern industry. Although fault diagnosis methods based on deep learning have achieved high accuracy, most of them only extract features from a single domain. Methods based on a single domain are difficult to apply to environments with noise. This paper presented a diagnosis method based on the attention mechanism and the fusion of time domain and frequency domain features to improve diagnosis accuracy. The presented method contains three modules. Firstly, two shallow convolutional neural networks are employed to extract the time domain and frequency domain features from the vibration signal. Then, the attention mechanism is adopted to extract important features and perform preliminary feature fusion. Finally, a deep convolutional network is used to fuse feature further and extract high-level features. The presented method can effectively fuse multi-domain features and improve diagnosis accuracy. This paper validates the effectiveness of the proposed method through a fault diagnosis experiment. A comparative experiment illustrates that the presented method has obvious advantages in noise resistance. When the signal to noise ratio equals 0dB, the diagnosis accuracy of the presented method is up to 6.4% higher than that of the single domain method.

Significance: Pulmonary arterial hypertension (PAH) is a progressive disease arising from the narrowing of pulmonary arteries (PAs) resulting in high pulmonary arterial blood pressure and ultimately right ventricle (RV) failure. A defining characteristic of PAH is the excessive and unrelenting inward remodeling of PAs that includes increased proliferation, inflammation, and fibrosis. Critical Issues: There is no cure for PAH nor interventions that effectively arrest or reverse PA remodeling, and intensive research over the past several decades has sought to identify novel molecular mechanisms of therapeutic value. Recent Advances: Galectin-3 (Gal-3) is a carbohydrate-binding lectin remarkable for its chimeric structure, composed of an N-terminal oligomerization domain and a C-terminal carbohydrate-recognition domain. Gal-3 has been identified as a regulator of numerous changes in cell behavior that contributes to aberrant PA remodeling, including cell proliferation, inflammation, and fibrosis, but its role in PAH has remained poorly understood until recently. In contrast, pathological roles for Gal-3 have been proposed in cancer and inflammatory and fibroproliferative disorders, such as pulmonary vascular and cardiac fibrosis. Herein, we summarize the recent literature on the role of Gal-3 in the development of PAH. We provide experimental evidence supporting the ability of Gal-3 to influence reactive oxygen species production, NADPH oxidase enzyme expression, and redox signaling, which have been shown to contribute to both vascular remodeling and increased pulmonary arterial pressure. Future Directions: While several preclinical studies suggest that Gal-3 promotes hypertensive pulmonary vascular remodeling, the clinical significance of Gal-3 in human PAH remains to be established. Antioxid. Redox Signal. 00, 000–000.

Perovskite solar cells (PSCs) have attracted increasing attentions for their outstanding efficiency and simple fabrication process. However, the commonly used typical titanium dioxide (TiO2) electron transport layer (ETL) requires a high-temperature treatment to crystallize. In this work, the low-temperature (150 °C) atomic layer deposition (ALD) processed TiO2 ETL, and the SnO2/PCBM double interfacial modifier have been adopted to fabricate the high performance PSCs of ITO/ALD-TiO2/SnO2/PC61BM/MA0.7FA0.3PbI(3-x)Cl(x)/Spiro-OMeTAD/Ag. The measurement results show that the PSCs with ALD-TiO2 ETL obtain the higher champion PCE of 19.45% than that of PSC with solution-TiO2 ETLs (PCE = 18.64%), along with the obviously enhanced FF and VOC, as well as the better light stability and repeatability. And the transient photocurrent and photovoltage results demonstrate that the performance improvement origins from the enhanced charge transport and suppressed charge recombination. Therefore, the low-temperature ALD-TiO2 ETL with suitable modification is a promising strategy to fabricate efficient PSCs and flexible devices.

A defining characteristic of pulmonary hypertension (PH) is the extensive remodeling of pulmonary arteries (PAs), which results in progressive increases in vascular resistance and stiffness and eventual failure of the right ventricle. There is no cure for PH and identification of novel molecular mechanisms that underlie increased proliferation, reduced apoptosis, and excessive extracellular matrix production in pulmonary artery smooth muscle cells (PASMCs) is a vital objective. Galectin-3 (Gal-3) is a chimeric lectin and potent driver of many aspects of fibrosis, but its role in regulating PASMC behavior in PH remains poorly understood. Herein, we evaluated the importance of increased Gal-3 expression and signaling on PA vascular remodeling and cardiopulmonary function in experimental models of PH. Gal-3 expression was quantified by qRT-PCR, immunoblotting, and immunofluorescence imaging, and its functional role was assessed by specific Gal-3 inhibitors and CRISPR/Cas9-mediated knockout of Gal-3 in the rat. In rat models of PH, we observed increased Gal-3 expression in PASMCs, which stimulated migration and resistance to apoptosis, whereas silencing or genetic deletion reduced cellular migration and PA fibrosis and increased apoptosis. Gal-3 inhibitors attenuated and reversed PA remodeling and fibrosis, as well as hemodynamic indices in monocrotaline (MCT)-treated rats in vivo. These results were supported by genetic deletion of Gal-3 in both MCT and Sugen Hypoxia rat models. In conclusion, our results suggest that elevated Gal-3 levels contribute to inappropriate PA remodeling in PH by enhancing multiple profibrotic mechanisms. Therapeutic strategies targeting Gal-3 may be of benefit in the treatment of PH.

A novel miniaturised antenna with a split elliptical patch and three left/right handed (LRH) cells is proposed. The antenna has a wideband circularly polarised (CP) characteristic and it is easy to perform the impedance matching. A simple microstrip line is used to excite the antenna and another microstrip line is matched by a proper chip resistor. By splitting the elliptical patch and adjusting the position of three LRH cells, an improved CP characteristic and a wide impedance bandwidth are achieved. Compared to the referenced antenna, the lower resonant frequency is decreased and the higher resonant frequency is increased. Also, the miniaturisation, wideband width and broadband CP are achieved. The measured results demonstrate that this structure achieves 69% (2.4-4.93 GHz) of the 3-dB axial ratio bandwidth and 107.9% (1.8-6.23 GHz) of the 10 dB return loss bandwidth. The antenna performs an average gain of over 2.7 dB across the CP region.

Accurate and timely prediction of remaining useful life (RUL) of a machine enables the machine to have an appropriate operation and maintenance decision. Data-driven RUL prediction methods are more attractive to researchers because they can be deployed quicker and cheaper compared to other approaches. The existing deep neural network (DNN) models proposed for the applications of RUL prediction are mostly single-path and top-down propagation. In order to improve the prognostic accuracy of the network, this paper proposes a directed acyclic graph (DAG) network that combines long short term memory (LSTM) and a convolutional neural network (CNN) to predict the RUL. Different from the existing prediction models combined with CNN and LSTM, the method proposed in this paper combines CNN and LSTM organically instead of just using CNN for feature extraction. Moreover, when a single timestamp is used as an input, padding the signals in the same training batch would affect the prediction ability of the developed model. To overcome this drawback, the proposed method generates a short-term sequence by sliding the time window (TW) with one step size. In addition, based on the degradation mechanism, the piece-wise RUL function is used instead of the traditional linear function. In the experimental test, the turbofan engine degradation simulation dataset provided by NASA is used to validate the proposed RUL prediction model. By comparing with the existing methods using the same dataset, it can be concluded that the prediction method proposed in this paper has better prediction capability.

Bifacial photovoltaic technology has been established as an effective strategy for increasing electric power generation while reducing the area-related costs of silicon solar cells. For third-generation solar cells of organolead halide perovskites, the monofacial power conversion efficiency (PCE) has exceeded 23%. Therefore, the development of efficient bifacial cells is expected, but the top transparent electrode is a crucial obstacle that must be overcome. In this work, ultrathin silver (Ag) film was selected as the top transparent electrode and, particularly, a tellurium oxide (TeO2) optical coupling layer was adopted to enhance the transmittance of the Ag film in the visible light region, which markedly increased the light absorption of devices. Compared with PSCs with a pure Ag electrode, the device with a TeO2 (40 nm)/Ag (11 nm) top electrode showed significantly enhanced bifacial short-circuit current-density and bifacial PCEs of 18.86% (illuminated from the ITO electrode) and 15.12% (illuminated from the Ag electrode). Consequently, in addition to its role as a transmittance enhancement layer, TeO2 acted as a back reflector when the device was illuminated from the ITO side. Furthermore, a previously proposed phenylethylammonium iodide treatment method was adopted to improve charge transport at the perovskite/fullerene interface. Remarkably, the device achieved champion bifacial PCEs of 20.96% (ITO side) and 17.36% (Ag side), a bifaciality factor of 82.82%, and a steady PCE output and good reproducibility. The large-area device (1 cm2) achieved bifacial PCEs of 12.42% (ITO side) and 10.55% (Ag side), with better performance predicted by simulations. Therefore, the TeO2/Ag films are promising transparent electrodes for fabricating cost-effective bifacial PSCs.

The quality of the perovskite layer plays a key role in the performance of perovskite solar cells (PSCs). In this work, we demonstrate a facile way to improve the quality of perovskite films by using a MAI/IPA solution to post-treat the perovskite material after the film formation. XRD, SEM and AFM show that an optimal, uniform, dense and highly crystalline perovskite film is achieved after the post-treatment. The transient photocurrent and transient photovoltage also confirm that the devices after the post-treatment have both the outstanding ability of charge extraction and suppressed charge recombination. By using the post-treatment, a maximum PCE of 19.08%, which is 12% higher than that of the control device, is obtained with a Jsc of 23.12 mA cm-2, a Voc of 1.08 V and an FF of 76.03%. The results show that the MAI/IPA solution post-treatment after the perovskite film formation is a facile way to improve the film quality and the PSC performance.

Ultrasonic guided wave has received significant attention in the field of nondestructive testing because of their advantages in detecting the entire wall thickness. Since pipe elbows are an important component of the piping system, it is desirable to study the propagation behavior of the guided wave in the elbows. In this paper, instead of using the traditional method of a ring of transducers to generate guided wave at the same time, the propagation behavior of the guided wave in the elbows under the action of a single excitation source is simulated. Using the chirp excitation signal, the transmission signals of the guided wave at different excitation central frequencies are extracted with high efficiency, and the behavior of the guided wave in the elbow under different excitation frequencies is studied. In addition, it is proposed to use the flight time quantity to characterize the complete information of the structure, and the influence of the defects on the flight time is also studied. The results prove that the flight time feature has good detection sensitivity to the defects with varying depth in different locations of the elbow.

This paper addresses the task of unsupervised video multi-object segmentation. Current approaches follow a two-stage paradigm: 1) detect object proposals using pre-trained Mask R-CNN, and 2) conduct generic feature matching for temporal association using re-identification techniques. However, the generic features, widely used in both stages, are not reliable for characterizing unseen objects, leading to poor generalization. To address this, we introduce a novel approach for more accurate and efficient spatio-temporal segmentation. In particular, to address \textbf{instance discrimination}, we propose to combine foreground region estimation and instance grouping together in one network, and additionally introduce temporal guidance for segmenting each frame, enabling more accurate object discovery. For \textbf{temporal association}, we complement current video object segmentation architectures with a discriminative appearance model, capable of capturing more fine-grained target-specific information. Given object proposals from the instance discrimination network, three essential strategies are adopted to achieve accurate segmentation: 1) target-specific tracking using a memory-augmented appearance model; 2) target-agnostic verification to trace possible tracklets for the proposal; 3) adaptive memory updating using the verified segments. We evaluate the proposed approach on DAVIS$_{17}$ and YouTube-VIS, and the results demonstrate that it outperforms state-of-the-art methods both in segmentation accuracy and inference speed., CVPR21

Gear pitting fault is one of the most common faults in mechanical transmission. Acoustic emission (AE) signals have been effective for gear fault detection because they are less affected by ambient noise than traditional vibration signals. To overcome the problem of low gear pitting fault recognition rate using AE signals and convolutional neural networks, this paper proposes a new method named augmented convolution sparse autoencoder (ACSAE) for gear pitting fault diagnosis using raw AE signals. First, the proposed method combines sparse autoencoder and one-dimensional convolutional neural networks for unsupervised learning and then uses the reinforcement theory to enhance the adaptability and robustness of the network. The ACSAE method can automatically extract fault features directly from the original AE signals without time and frequency domain conversion of the AE signals. AE signals collected from gear test experiments are used to validate the ACSAE method. The analysis result of the gear pitting fault test shows that the proposed method can effectively performing recognition of the gear pitting faults, and the recognition rate reaches above 98%. The comparative analysis shows that in comparison with fully-connected neural networks, convolutional neural networks, and recurrent neural networks, the ACSAE method has achieved a better diagnostic accuracy for gear fitting faults.

Research on data-driven fault diagnosis methods has received much attention in recent years. The deep belief network (DBN) is a commonly used deep learning method for fault diagnosis. In the past, when people used DBN to diagnose gear pitting faults, it was found that the diagnosis result was not good with continuous time domain vibration signals as direct inputs into DBN. Therefore, most researchers extracted features from time domain vibration signals as inputs into DBN. However, it is desirable to use raw vibration signals as direct inputs to achieve good fault diagnosis results. Therefore, this paper proposes a novel method by stacking spare autoencoder (SAE) and Gauss-Binary restricted Boltzmann machine (GBRBM) for early gear pitting faults diagnosis with raw vibration signals as direct inputs. The SAE layer is used to compress the raw vibration data and the GBRBM layer is used to effectively process continuous time domain vibration signals. Vibration signals of seven early gear pitting faults collected from a gear test rig are used to validate the proposed method. The validation results show that the proposed method maintains a good diagnosis performance under different working conditions and gives higher diagnosis accuracy compared to other traditional methods.

This paper deals with gear pitting fault diagnosis problem and presents a method by integrating convolutional neural network (CNN) and gated recurrent unit (GRU) networks with vibration and acoustic emission signals to solve the problem. The presented method first trains a one-dimensional CNN with acoustic emission signals and a GRU network with vibration signals. Then the gear pitting fault features obtained by the two networks are concatenated to form a deep learning structure for gear pitting fault diagnosis. Seven different gear pitting conditions are used to test the feasibility of the presented method. The diagnosis result of the gear pitting fault shows that the accuracy of the presented method reaches above 98% with only a relatively small number of training samples. In comparison with the results using CNN or GRU network alone, the presented method gives more accurate diagnosis results. By comparing the results of different loads and learning rates, the robustness of the presented method for gear pitting fault diagnosis is proved. Moreover, the presented deep structure can be easily extended to more other sensor input signals for gear pitting fault diagnosis in the future.

Optical microcavity configuration is one optical strategy to enhance light trapping in devices using planar electrodes. In this work, the potential application of optical microcavity configuration with ultrathin metal electrodes in highly efficient perovskite solar cells (PSCs) was investigated. By comparing with the device with conventional indium-tin-oxide (ITO) electrodes, it is shown that by carefully designing the Ag/dielectric planar electrode, a device with an optical microcavity structure can achieve comparable&mdash, or even higher&mdash, power conversion efficiency than a conventional device. Moreover, there is a relative high tolerance for the Ag film thickness in the optical microcavity structure. When the thickness of the Ag film is increased from 8 to 12 nm, the device still can attain the performance level of a conventional device. This gives a process tolerance to fabricate devices with an optical microcavity structure and reduces process difficulty. This work indicates the great application potential of optical microcavities with ultrathin metal electrodes in PSCs, more research attention should be paid in this field.

Medium-carbon steel is widely used in architecture, rail, machinable steel and so on. So, there is a huge significance in the analysis and research of its fatigue life. In this paper, the fatigue experiment with different surface roughness was set up. In the experiment, there were three type of roughness and a total of 75 experiments were performed. Then a deep autoencoder to model the relationship of roughness (R), stress (S), and fatigue life (N). It has been found that the method can automatically extract the features which can be used to effectively model the relationship of R, S and N. Finally, the approach presented was compared with the existing model based on Tanaka-Mura theory and got an unexpected better result.

Gear pitting fault is common in mechanical devices. At present, most of the gear pitting fault detection methods are based on the manual extraction of the frequency domain features from vibration signals. This paper presents a method for gear pitting fault level diagnosis using vibration signals with an improved inception network. The presented method directly applies to the vibration signals to automatically extract features and diagnose the level of the gear pitting fault using deep learning. The presented method has been validated with vibration data collected for 7 gear pitting conditions from gear pitting fault tests. The validation results have shown that the presented method can effectively classify the levels of the gear pitting faults. In comparison with traditional convolutional neural network, the diagnosis accuracy has been significantly improved with the presented method.

International audience; Reliable gas supply for minimum risk of supply shortage and minimum power demand for low energy cost are two fundamental objectives of natural gas pipeline networks. In this paper, a multi-objective optimization method is developed to trade-off reliability and power demand in the decision process. In the optimization, the steady state behavior of the natural gas pipeline networks is considered, but the uncertainties of the supply conditions and customer consumptions are accounted for. The multi-objective optimization regards finding operational strategies that minimize power demand and risk of gas supply shortage. To quantify the probability of supply interruption in pipeline networks, a novel limit function is introduced based on the mass conservation equation. Then, the risk of interruption is calculated by combining the probability of interruption and its consequences, measured in utility terms. The multi-objective optimization problem is solved by the NSGA-II algorithm and its effectiveness is tested on two typical pipeline networks, i.e., a tree-topology network and a loop-topology network. The results show that the developed optimization model is able to find solutions which effectively compromise the need of minimizing gas supply shortage risk and reducing power demand. Finally, a sensitivity analysis is conducted to analyze the impact of demand uncertainties on the optimization resu

For reference antenna without parasitic patch, it covers the impedance operations from 1.72 to 4.73 GHz and 5.38 to 6.62 GHz with | S 11 |≤-10 dB, and has dual sense circular polarisation (CP) between 3.4 and 3.8 GHz (11.1%), between5.5 and 5.63 GHz (2.3%), respectively. In this study, to further improve the performance, an enhanced coplanar waveguide fed dual CP broad bandwidth antenna is proposed by using the C-shaped parasitic patch. It covers the impedance operations from 1.72 to 6.6 GHz with | S 11 |≤-10 dB, the 3-dB axial ratio between 2.92 and 4.04 GHz (32.2%) for left-hand CP as well as that between 5.42 and 5.63 GHz (3.8%) for right-hand CP are achieved. Compared with reference antenna, it is obviously that the proposed antenna achieves a remarkably enhanced dual CP with wider resonant bandwidth. Furthermore, the measured results and simulated results are agreement well with each other.

Dopamine D3 receptor (D3R) on immune cells is involved in the pathogenesis of rheumatoid arthritis (RA). Mast cells (MCs) are currently identified as important effector cells in synovial inflammation of RA, but little is known about the role of D3R on synovial MCs in the pathogenesis of RA. Several inflammatory cells in the synovium induce reactive oxygen species (ROS) formation which are involved in the progression of RA. However, it is unclear whether D3R on synovial MCs is related to the levels of ROS in RA patients. In this study, a total of 73 patients with RA were divided into three groups according to disease activity DAS28 scores. The number of cases in group 1, group 2, and group 3 was 19, 26, and 28, respectively. We examined D3R-positive MC numbers in the synovial fluid and ROS levels in each group of RA patients, and we also analyzed the association of D3R-positive MC numbers with RA disease activity and ROS levels. MDA and protein carbonylation in the serum and synovial fluid were measured to reflect the level of lipid peroxidation and protein oxidation, respectively. Additionally, superoxide dismutase (SOD) and catalase (CAT) in the serum and synovial fluid were used to be markers of antioxidant levels. Our results showed that D3R-positive MCs in the synovial fluid showed a declining trend with the increased disease activity DAS28 score in RA patients. There was negative correlation between D3R-positive MC numbers in the synovial fluid and disease severity DAS28 score of RA patients. Moreover, D3R-positive MC numbers in the synovial fluid were negatively correlated with the level of MDA and protein carbonylation while were positively correlated with antioxidant levels such as SOD and CAT in RA patients. Our results suggested that D3R on MCs may be involved in ROS-mediated pathogenesis of RA.

Bifacial semitransparent inverted planar structured perovskite solar cells (PSCs) based on Cs0.05FA0.3MA0.7PbI2.51Br0.54 using an Ag thin film electrode and V2O5 optical coupling layer are investigated theoretically and experimentally. It is shown that the introduction of the cesium (Cs) ions in the perovskite could obviously improve the device performance and stability. When only the bare Ag film electrode is used, the PSCs show a bifacial performance with the power conversion efficiency (PCE) of 14.62% illuminated from the indium tin oxide (ITO) side and 5.45% from the Ag film side. By introducing a V2O5 optical coupling layer, the PCE is enhanced to 8.91% illuminated from the Ag film side, which is 63% improvement compared with the bare Ag film electrode, whereas the PCE illuminated from the ITO side remains almost unchanged. Moreover, when a back-reflector is employed, the PCE of device could be further improved to 15.39% by illumination from the ITO side and 12.44% by illumination from the Ag side. T...

The effect of chronic insomnia carries along a physical and psychological state of dysfunctionality felt by the individual. The purpose of the study was to investigate the effects of hypnotherapeutic intervention applied in the case of patients who present insomnia accompanied by rumination. We performed the study of 42 patients who underwent the Symptom Checklist-90 (SCL-90) for the psychiatric evaluation and the Pittsburgh Sleep Quality Index (PSQI) for sleep quality at baseline, 1 month after the program and follow-up (2 months) after finalizing the program. Our findings revealed that hypnotherapeutic intervention that involves relaxation and positive self-suggestion leads to positive results among the patient who experience insomnia accompanied by rumination.

A new kind of flexible and stretchable strip-shaped thermochromic resistive heater (TRH) has been fabricated by incorporating an aligned carbon nanotube sheet and a thermochromic silicone elastomer. This strip-shaped TRH demonstrates rapid thermal response and high stability even under stretching at a speed of 2 mm s−1. The resulting TRH textiles woven from the strip-shaped TRHs are flexible, stretchable, and breathable, and they can stably work under various deformations such as twisting. The temperatures of TRH textiles during working can be visually evaluated from the designed patterns for both high efficiency and safety. The weaving structure in the TRH textile is available for local heating, which has been demonstrated for thermal therapy.

The early gear pitting fault diagnosis has received much attention in the industry. In recent decades, with the popularity growth of artificial neural network, researchers have applied deep learning methods to figure out early gear pitting faults. However, the classical fault diagnosis methods usually use deep neural networks according to the time sequence of the collected signals. In this case, the feature extraction in the direction of the inverse time-domain signals is usually ignored. Aimed at overcoming this shortage, ground on a traditional Long Short Term Memory (LSTM) network, this paper proposes a Bidirectional LSTM (Bi-LSTM) to construct a fault diagnosis model of early gear pitting using raw vibration signals. Using the Bi-LSTM network, feature extraction of the vibrational signals in both directions is simultaneously carried out to evaluate the degree of the early gear pitting faults to better extract the gear pitting characteristics from the raw vibration signals of the gear. Through the analysis of the experimental data, compared with the traditional LSTM model, the Bi-directional LSTM has a classification accuracy of over 96% for early gear pitting fault diagnosis, which is an increase of 4.1%.

Prognostics and health management (PHM) has received much attention as an emerging discipline. And the prediction of remaining useful life (RUL) is the core of the PHM. The data-driven RUL prediction methods are more favored because they can be developed faster and cheaper. However, the existing data-driven prediction models usually can only be used under the same data domain (DD), and they require a lot of labeled data to retrain a new prediction model. So a domain adaptation prediction model is more desirable. In this paper, a domain adaptation RUL prediction model is proposed by integrating the adaptive batch normalization (AdaBN) into deep convolutional neural network (DCNN). The improved AdaBN-DCNN model can not only improve the accuracy of the prediction, but also adapt to the prognostic tasks under different DDs. The sliding time window (TW) and the improved piecewise linear RUL function are also used in this paper to improve the prediction capability of the model. The proposed RUL prediction model is validated using the C-MAPSS turbofan engine dataset provided by NASA. The prediction results show that the proposed model not only has a strong predictive power but also adapts to different DDs.

International audience; The rapid development of big data and smart technology in the natural gas industry requires timely and accurate forecasting of natural gas consumption on different time horizons. In this work, we propose a robust hybrid hours-ahead gas consumption method by integrating Wavelet Transform, RNN-structured deep learning and Genetic Algorithm. The Wavelet Transform is used to reduce the complexity of the forecasting tasks by decomposing the original series of gas loads into several sub-components. The RNN-structured deep learning method is built up via combining a multi-layer Bi-LSTM model and a LSTM model. The multi-layer Bi-LSTM model can comprehensively capture the features in the sub-components and the LSTM model is used to forecast the future gas consumption based on these abstracted features. To enhance the performance of the RNN-structured deep learning model, Genetic Algorithm is employed to optimize the structure of each layer in the model. Besides, the dropout technology is applied in this work to overcome the potential problem of overfitting. In this case study, the effectiveness of the developed method is verified from multiple perspective, including graphical examination, mathematical errors analysis and model comparison, on different data sets.

International audience; Advanced sensor and communication technologies can make natural gas supply systems smarter than ever before, in both system management and operation. This paper presents the development of a novel data-driven Demand Side Management, whose framework includes demand forecasting, customer response analysis, prediction of dynamic condition of the gas network, quick supply reliability evaluation, multi-objective optimization and decision-making. The aims of this DSM method are to smooth load profiles, improve company profit and enhance system reliability, by means of a dynamic pricing strategy. To verify the effectiveness of the developed framework, a case study is considered, concerning the management of a relatively complex gas supply system, wherein four different pricing periods are introduced for comprehensively testing. The results in the case study show that the DSM framework is able to effectively achieve the targets of peak shaving and valley filling. Besides, it can significantly and stably improve the system efficiency and reliability, for different pricing periods. Finally, pricing period determination is discussed in relation to the features of performance.

As a multimedia security mechanism, CAPTCHAs are completely automated public turing test to tell computers and humans apart. Although cracking CAPTCHA has been explored for many years, it is still a challenging problem for real practice. In this demo, we present a text based CAPTCHA cracking system by using convolutional neural networks(CNN). To solve small sample problem, we propose to combine conditional deep convolutional generative adversarial networks(cDCGAN) and CNN, which makes a tremendous progress in accuracy. In addition, we also select multiple models with low pearson correlation coefficients for majority voting ensemble, which further improves the accuracy. The experimental results show that the system has great advantages and provides a new mean for cracking CAPTCHAs.

Simultaneous wireless information and power transfer (SWIPT) is a potential technique to solve the energy scarcity problem in wireless communication networks. Information transmission security is one of the key issues to be addressed in SWIPT. In this study, the authors consider guaranteeing the information transmission security of a multiple-input multiple-output non-regenerative relay system with SWIPT in the physical layer by designing source and relay beamforming to maximise the security rate of the system. The problem is to design the source signal and artificial noise (AN) beamforming and the relay forwarding beamforming jointly, which is highly non-convex. They first propose an alternating optimisation-based algorithm, which designs the beamforming matrices at the source and relay separately and alternately. Next, they propose a joint design algorithm which designs the source signal and AN beamforming and the relay beamforming together. Finally, they propose a non-iterative design algorithm with low complexity. The performances of the proposed design algorithms have been verified by computer simulations.

Cognitive radio is a promising technique to solve the spectrum shortage problem in wireless communications. Integrating wireless relaying into the cognitive radio networks can improve the spectrum efficiency. In this paper, a multi-user multiple-input multiple-output cognitive multi-relay network is considered, where the primary network (PN) consists of one transmitter-receiver pair and the secondary network (SN) consists of multiple active source–destination pairs and multiple non-regenerative relays. All nodes in the networks are deployed with multiple antennas. Transceivers in SN should suppress the interference from SN to PN and guarantee the quality of SN transmission simultaneously. We deal with the transceiver design problem with practical condition that the system only has imperfect channel state information (CSI). The transceivers in SN, i.e. source precoders, relay beamforming and destination receiving filters, are jointly designed to minimize the sum mean square error (MSE) subject to interference power constraints and individual transmit power constraints. We propose an algorithm to divide the design problem into three subproblems, which can be solved using the convex optimization methods. The algorithm solves them iteratively until convergence. The performances of the proposed design algorithm have been verified by computer simulations.

A novel polymer-based optoelectronic device is created to independently realize either photoelectric conversion or light emission. It is produced by coating photoactive materials on one side of an aligned carbon nanotube sheet electrode and polymer light-emitting materials on the other through an all-solution process. The power conversion efficiency exceeds 2% and the brightness reaches 1040 cd m−2 for photoelectric conversion and light emission, respectively. This study indicates a promising potential for the all-solution-based process for a wide variety of applications, including microelectronics, and smart and multi-function devices.

Gears are the most common parts of a mechanical transmission system. Gear wearing faults could cause the transmission system to crash and give rise to the economic loss. It is always a challenging problem to diagnose the gear wearing condition directly through the raw signal of vibration. In this paper, a novel method named augmented deep sparse autoencoder (ADSAE) is proposed. The method can be used to diagnose the gear wearing fault with relatively few raw vibration signal data. This method is mainly based on the theory of wearing fault diagnosis, through creatively combining with both data augmentation ideology and the deep sparse autoencoder algorithm for the fault diagnosis of gear wear. The effectiveness of the proposed method is verified by experiments of six types of gear wearing conditions. The results show that the ADSAE method can effectively increase the network generalization ability and robustness with very high accuracy. This method can effectively diagnose different gear wearing conditions and show the obvious trend according to the severity of gear wear faults. This paper provides an important insight into the field of gear fault diagnosis based on deep learning and has a potential practical application value.

Objective: To explore the expression and significance of CD28, ctla-4, CD80 and CD86 in ankylosing spondylitis. Methods: Flow cytometry was used to test in January 2016-January 2017 in our hospital was 73 cases of ankylosing spondylitis patients and 40 normal controls CD28, CTLA 4, CD80 and CD86 weeks outside expression in lymphocytes. By ELISA method, the lgA serum immunoglobulin IgG, IgM and hypersensitive c-reactive protein (hs CRP) and blood sedimentation (ESR) levels, and to explore, CD80, CD86 and CD28, CTLA - 4, age, duration of the hs CRP, ESR, Bath AS functional index (BASFI) and Bath AS measurement index (BASMI) relevance. Results: The levels of CD28, ctla-4, CD80 and CD86 were significantly higher in patients with ankylosing spondylitis (P

International audience; A systematic method is developed for supply reliability assessment of natural gas pipeline networks. In thedeveloped method, the integration of stochastic processes, graph theory and thermal-hydraulic simulation isperformed accounting for uncertainty and complexity. The supply capacity of a pipeline network depends on theunit states and the network structure, both of which change stochastically because of stochastic failures of theunits. To describe this, in this work a capacity network stochastic model is developed, based on Markov mod-eling and graph theory. The model is embedded in an optimization algorithm to compute the capacities of thepipeline network under different scenarios and analyze the consequences of failures of units in the system.Indices of supply reliability and risk are developed with respect to two aspects: global system and individualcustomers. In the case study, a gas pipeline network is considered and the results are analyzed in detail.

International audience; A systematic framework is developed to assess the vulnerability of natural gas pipeline networks. To measure the impact of accidents on gas supply service, a consequence model is developed based on a flow algorithm embedded into an optimization scheme, with the consideration of physical constraints. The vulnerability analysis is performed from three viewpoints: global vulnerability analysis, demand robustness and critical pipeline analysis. The global vulnerability analysis is performed considering hazards and threats in gas sources, demand and transmission system. The analysis of demand robustness evaluates the capacities of demand sites to withstand the strains imposed on the pipeline network system and further explains the differences in capacities, from a graph theory perspective. In the critical pipeline analysis, criticalities of pipelines are evaluated by considering direct attacks and using a physical flow-based method. The analyses are performed on a relatively complex gas pipeline network taken from literature.

International audience; The current framework of management of natural gas pipeline systems, based on off-line simulation, is facing challenges because of the increasing complexity, uncertainty and a number of time-dependent factors. To be effective, it requires comprehensive knowledge of system characteristics, accurate initial and boundary conditions. In an attempt to circumvent these problems, in this work we propose to use the deep learning method in the natural gas transmission system operation and management context. A data-driven prediction method is developed from real-time data of operation pressure and gas consumption. Specifically, the deep learning method is combined with the data window method and structural controllability theory to predict the conditions of gas pipeline network components. The data window method is applied to reconstruct the data structure and build a "memory" for the deep learning method. Structural controllability theory is applied to extract critical parameters, for reducing the problem size. The developed method allows accurate and efficient predictions, especially in abnormal conditions. For demonstration, the method is applied to a complex gas pipeline network. The results show that the developed method can provide accurate real-time predictions useful for reducing potential losses in operation, and perform efficient and effective management of the gas pipeline system. In the case study, the average prediction accuracy is higher than 0.99.

The transmission of spur gear mechanism may generate uneven load distribution because of machining error, assembly error, elastic deformation, and other factors, resulting in serious damage to the bearing capacity and service life of gears. To improve the contact condition of gear pairs and enhance the meshing performance and bearing capacity, this study proposed a novel axial modification method based on a composite modification curve with indefinite parameters and an evaluation method to evaluate modification effects. In addition to the surface equation of modified tooth was derived according to the composite modification curve, and the finite element model was built for gear pair. After conducting simulation analysis for the meshing process, the location and the shape of contact area as well as the other results can be acquired. In addition, the modification parameters can be optimized by performing orthogonal experiments for modified gear pairs; thus, the ideal modification effect is obtained at a specific operating condition. Moreover, comparison analysis was performed, and the results show that the phenomenon of uneven load distribution is dramatically improved when the gear pair was modified with optimized parameters, and the bearing capacity of the gear pair was increased. Finally, using the optimized parameters to trial-produce gears and conducting running-in tests, the effectiveness and the practicability of the method proposed in this study were verified.

This paper presents finite element analysis of the strength of roadheaher conical picks. Failure forms of conical picks are investigated, and the FE model of conical pick is established which is based on ANSYS. The static analysis is carried out according to the load calculated by empirical formula and the deformation map and stress distribution is obtained. The results show that the maximum stress of conical pick appeared in the tool bit. In order to examine the result of static analysis, contact analysis of conical pick cutting process is researched, and a consistency is achieved. The finite element analysis not only offers a new way of studying the strength of conical picks and improving the service life of the pick, but also provides a reliable theoretical basis for the selection, design and use of the conical picks.

Enthesitis is considered as the primary anatomical lesion in ankylosing spondylitis (AS). We aimed to investigate the potential of ultrasound to detect early changes after TNF-a antagonist therapy of Achilles enthesitis of AS patients. One hundred AS patients with active disease, requiring TNF-a antagonist therapy, were included (etanercept n = 25, infliximab n = 25, adalimumab n = 25, non-biologic disease-modifying antirheumatic drugs (DMARDs) n = 25). Physical examination was performed to evaluate disease activity and detect Achilles enthesitis and/or retrocalcaneal bursitis. Ultrasound of the Achilles enthesitis was performed bilaterally. Follow-up examinations were performed 3 months after the initiation of therapy. Gray scale (GS) scores, Power Doppler (PD) scores, and total additive scores (TS) decreased significantly during TNF-a antagonist therapy but not in traditional non-biologic traditional DMARDs group. The bath ankylosing spondylitis disease activity index (BASDAI), bath ankylosing spondylitis metrology index (BASMI), bath ankylosing spondylitis functional index (BASFI), and Maastricht ankylosing spondylitis enthesitis score (MASES) all showed significant improvements. When three different TNF-a antagonists were analyzed separately, no significant difference was observed in GS, PD, and total scores. Subclinical Achilles enthesitis, detected only with GS ultrasound, is present in a subset of AS patients and a significant improvement can be demonstrated after 3 months of TNF-a antagonist therapy. Doppler ultrasound provides a reliable estimation to monitor the therapeutic response to TNF antagonists in AS patients with Achilles enthesitis. TNF-a antagonists have been shown to be effective in decreasing ultrasound signs of enthesitis after 3 months of therapy in AS patients.