Partial Transfer Ensemble Learning Framework: A Method for Intelligent Diagnosis of Rotating Machinery Based on an Incomplete Source Domain.

Most cross-domain intelligent diagnosis approaches presume that the health states in training datasets are consistent with those in testing. However, it is usually difficult and expensive to collect samples under all failure states during the training stage in actual engineering; this causes the training dataset to be incomplete. These existing methods may not be favorably implemented with an incomplete training dataset. To address this problem, a novel deep-learning-based model called partial transfer ensemble learning framework (PT-ELF) is proposed in this paper. The major procedures of this study consist of three steps. First, the missing health states in the training dataset are supplemented by another dataset. Second, since the training dataset is drawn from two different distributions, a partial transfer mechanism is explored to train a weak global classifier and two partial domain adaptation classifiers. Third, a particular ensemble strategy combines these classifiers with different classification ranges and capabilities to obtain the final diagnosis result. Two case studies are used to validate our method. Results indicate that our method can provide robust diagnosis results based on an incomplete source domain under variable working conditions.

Fusion Domain-Adaptation CNN Driven by Images and Vibration Signals for Fault Diagnosis of Gearbox Cross-Working Conditions.

The vibration signal of gearboxes contains abundant fault information, which can be used for condition monitoring. However, vibration signal is ineffective for some non-structural failures. In order to resolve this dilemma, infrared thermal images are introduced to combine with vibration signals via fusion domain-adaptation convolutional neural network (FDACNN), which can diagnose both structural and non-structural failures under various working conditions. First, the measured raw signals are converted into frequency and squared envelope spectrum to characterize the health states of the gearbox. Second, the sequences of the frequency and squared envelope spectrum are arranged into two-dimensional format, which are combined with infrared thermal images to form fusion data. Finally, the adversarial network is introduced to realize the state recognition of structural and non-structural faults in the unlabeled target domain. An experiment of gearbox test rigs was used for effectiveness validation by measuring both vibration and infrared thermal images. The results suggest that the proposed FDACNN method performs best in cross-domain fault diagnosis of gearboxes via multi-source heterogeneous data compared with the other four methods.

Job burnout and its influencing factors among village doctors during the COVID-19 pandemic: a cross-sectional study.

Objective: The aim of this study is to understand the job burnout of village doctors during the COVID-19 epidemic and its influencing factors, and to provide a reference for effectively alleviating the job burnout of village doctors. Methods: A cross-sectional survey was conducted among village doctors in S province in December 2021. The survey included a general information questionnaire and the CMBI Burnout Scale. Epidata was used for dual input, and descriptive analysis, t-test, chi-square test, and binary Logistic regression for statistical analysis were used. Results: A total of 993 village doctors participated in the survey. Most of them were male village doctors (62.84%), with an average age of 46.57 (SD = 7.50). Village doctors believed that the impact of the epidemic on work was serious, with a score of 3.87 ± 0.91. The economic support was small, with a score of 2.31 ± 0.99. The development space was low, with a score of 2.62 ± 0.98. The overall incidence of burnout was 53.47%. In the burnout group, 54.05% were mild, 33.14% were moderate, and 12.81% were severe. The high degree of difficulty in using WeChat (OR = 1.436, 95%CI: 1.229-1.679), high work pressure (OR = 1.857, 95%CI: 1.409-2.449), high risk of practice (OR = 1.138, 95%CI: 1.004-1.289), less economic support (OR = 0.825, 95%CI: 0.684-0.995), less technical support (OR = 0.696, 95%CI: 0.565-0.858), and poor emotional support (OR = 0.632, 95%CI: 0.513-0.780) were more likely to have job burnout. Conclusion: Burnout is a common phenomenon among village doctors during the COVID-19 pandemic, which needs to be prevented and alleviated by various measures.

[Multi-spectral thermometry based on radiation measurement within a finite solid-angle].

Based on Planck's law, the surface temperature of an object can be determined by measurement of emitted radiation. The equation for monochromatic radiation thermometry within a finite solid-angle was deduced, and it was found that if the surface temperature and spectral emissivity can be solved at the same time, the specific radiation measurement conditions for multi-spectral thermometry should be generally met that the radiation measurement should be implemented within an infinitesimal solid-angle or within a finite solid-angle only for a perfect diffuser. When the directional spectral emissivity modeled by finite polynomial series is employed and proper mathematical transformation is used, a universal equation for monochromatic radiation thermometry is obtained. So the restrictions in radiation measurement can be got rid of, but spectral emissivity may not be solved simultaneously. Multi-solution problem was preliminarily investigated, and so a solution was put forward that the channel number should be more than the number of the variables to be solved and the nonlinear least squares method should be used.

Identification and validation of diagnostic biomarkers of coronary artery disease progression in type 1 diabetes via integrated computational and bioinformatics strategies.

OBJECTIVE: Our study aimed to identify early peripheral blood diagnostic biomarkers and elucidate the immune mechanisms of coronary artery disease (CAD) progression in patients with type 1 diabetes mellitus (T1DM). METHODS: Three transcriptome datasets were retrieved from the Gene Expression Omnibus (GEO) database. Gene modules associated with T1DM were selected with weighted gene co-expression network analysis. Differentially expressed genes (DEGs) between CAD and acute myocardial infarction (AMI) peripheral blood tissues were identified using limma. Candidate biomarkers were selected with functional enrichment analysis, node gene selection from a constructed protein-protein interaction (PPI) network, and 3 machine learning algorithms. Candidate expression was compared, and the receiver operating characteristic curve (ROC) and nomogram were constructed. Immune cell infiltration was assessed with the CIBERSORT algorithm. RESULTS: A total of 1283 genes comprising 2 modules were detected as the most associated with T1DM. In addition, 451 DEGs related to CAD progression were identified. Among them, 182 were common to both diseases and mainly enriched in immune and inflammatory response regulation. The PPI network yielded 30 top node genes, and 6 were selected using the 3 machine learning algorithms. Upon validation, 4 genes (TLR2, CLEC4D, IL1R2, and NLRC4) were recognized as diagnostic biomarkers with the area under the curve (AUC) > 0.7. All 4 genes were positively correlated with neutrophils in patients with AMI. CONCLUSION: We identified 4 peripheral blood biomarkers and provided a nomogram for early diagnosing CAD progression to AMI in patients with T1DM. The biomarkers were positively associated with neutrophils, indicating potential therapeutic targets.

[Radiation thermometry based on calibration of spectral responsivity].

Abstract True surface temperatures can be determined by measurements of radiation emitted by the object. The non-spectral parameter in the radiation measurement equation is the function of the relative position between the target and the lens, so calibration of space position is necessary for temperature measurement, when emissivity and temperature are measured simultaneously. In the present paper, the non-spectral parameter was included into the undetermined coefficients of emissivity modeled by finite series, which will not affect the solution of true surface temperature. Therefore, radiation thermometry can be accomplished without calibration of space position and normalization of measurement data. And not the true spectral emissivity but the trend of it can be measured. Two special examples were investigated, respectively. The results indicate that when the effective wavelength of each channel is different, multi-wavelength radiation thermometry equations have the unique solution, while the number of the multiband ones may be zero, one, two or even three.

Nitrogen-doped mesoporous carbon single crystal-based Ag nanoparticles for boosting mild CO2 conversion with terminal alkynes.

Fabrication of efficient heterogeneous catalysts with high turnover frequency (TOF) is intriguing for rapid and scalable CO2 conversion under mild conditions, but it still faces some challenges due to use of some bulky and irregular supports causing inaccessible inner pores and insufficient utilization of active sites. Herein, using a unique nitrogen-doped mesoporous single-crystal carbon (named IRFC) as a host for loading Ag nanoparticles for the first time, a series of Ag/IRFC catalysts with high TOF (8.7-22.3 h-1) were facilely prepared by a novel "impregnation and in-situ reduction" strategy. The neat morphology and high porosity of IRFC with abundant N species, providing homogeneous surface, adequate space and anchoring sites for Ag immobilization, greatly facilitated the formation of highly-distributed ultrasmall Ag nanoparticles (2.3 nm). Meanwhile, smooth and short diffusion pathways were inherited from the ordered mesopores and small particle sizes of IRFC. Owing to these unparalleled structural features, the Ag/IRFC catalysts exhibited excellent catalytic activity, stability, and generality for mild CO2 conversion even under diluted conditions. This work not only presents a novel catalyst for mild CO2 conversion, but also brings some inspirations to designing highly efficient catalysts using well-shaped supporting nanomaterials for direct utilization of low-concentration CO2, such as flue gas.

Facilely controlled synthesis of a core-shell structured MOF composite and its derived N-doped hierarchical porous carbon for CO2 adsorption.

A new strategy for controlled synthesis of a MOF composite with a core-shell structure, ZIF-8@resorcinol-urea-formaldehyde resin (ZIF@RUF), is reported for the first time through in situ growth of RUF on the surface of ZIF-8 nanoparticles via an organic-organic self-assembly process by using hexamethylenetetramine as a formaldehyde-releasing source to effectively control the formation rate of RUF, providing the best opportunity for RUF to selectively grow around the nucleation seeds ZIF-8. Compared with the widely reported method for MOF composite synthesis, our strategy not only avoids the difficulty of incorporating MOF crystals into small pore sized materials because of pore limitation, but also effectively guarantees the formation of a MOF composite with a MOF as the core. After carbonization, a morphology-retaining N-doped hierarchical porous carbon characterized by its highly developed microporosity in conjunction with ordered mesoporosity was obtained. Thanks to this unique microporous core-mesoporous shell structure and significantly enhanced porosity, simultaneous improvements of CO2 adsorption capacity and kinetics were achieved. This strategy not only paves a way to the design of other core-shell structured MOF composites, but also provides a promising method to prepare capacity- and kinetics-increased carbon materials for CO2 capture.

Potassium Tethered Carbons with Unparalleled Adsorption Capacity and Selectivity for Low-Cost Carbon Dioxide Capture from Flue Gas.

Carbons are considered less favorable for postcombustion CO2 capture because of their low affinity toward CO2, and nitrogen doping was widely studied to enhance CO2 adsorption, but the results are still unsatisfactory. Herein, we report a simple, scalable, and controllable strategy of tethering potassium to a carbon matrix, which can enhance carbon-CO2 interaction effectively, and a remarkable working capacity of ca. 4.5 wt % under flue gas conditions was achieved, which is among the highest for carbon-based materials. More interestingly, a high CO2/N2 selectivity of 404 was obtained. Density functional theory calculations evidenced that the introduced potassium carboxylate moieties are responsible for such excellent performances. We also show the effectiveness of this strategy to be universal, and thus, cheaper precursors can be used, holding great promise for low-cost carbon capture from flue gas.