Contrast-Assisted Domain-Specificity-Removal Network for Semi-Supervised Generalization Fault Diagnosis.

Unknown domain shift caused by the unavailability of target domain during training phase degrades the performance of intelligent fault diagnosis models in practical applications. Domain generalization (DG)-based methods have recently emerged to alleviate the influence of domain shift and improve the generalization ability of models toward invisible working conditions. However, most existing studies are conducted on multiple fully labeled source domains. Meanwhile, domain-specific information related to the variations of working conditions is often neglected during model training. Therefore, in order to realize reliable generalization fault diagnosis based on partially labeled source domains, this article proposes a contrast-assisted domain-specificity-removal network (CDSRN) to extract transferable features from domain-specificity-removal perspective. Concretely, a domain-specific feature removal branch is designed to disentangle domain-invariant features and domain-specific features, thus excavating generalized information only in domain-invariance dimension. Simultaneously, proxy-contrastive representation enhancement module is embedded to facilitate the fault class-discriminative and domain-discriminative feature learning, thereby assisting the model in further improvement of generalization capability. Experimental studies confirm the effectiveness and competitiveness of the proposed CDSRN in semi-supervised generalization fault diagnosis.

[The application of network scale-up method on female sex workers and clients size estimation in Taizhou city].

OBJECTIVE: To estimate the size of female sex workers and clients in Taizhou city. METHODS: A household survey using network scale-up method (NSUM) was conducted among the 3000 community residents in Taizhou city from August to October in 2011, which aimed to estimate the social network size (c value) of Taizhou residents, and the c value was adjusted by demographic characteristics, back estimation and outlier elimination. Using the adjusted c value, the number of acquaintance of female sex workers or clients and the respect level toward female sex workers or clients were used to estimate the size of female sex workers and clients. RESULTS: A total of 2783 valid questionnaires were collected, among which 1380 (49.6%) were collected from Taixing city, 1403 (50.4%) were collected from Jingjiang city. 1334 respondents were male (47.9%) and 1449 (47.9%) respondents were female. The mean age was (39.4 ± 10.7) years. The average personal social network size using original data for Taizhou residents was 525, which differed from place, sex, age, educational level and marriage status. Using the remaining known populations through back estimation, the social network size was 419 and became 424 after the elimination of outliers. The estimated population size for female sex worker was 6370 (95%CI: 5886 - 6853), which accounted for 0.52% (6370/1 229 980) of the total number of female aged from 15 to 49. The estimated population size for clients was 15 202 (95%CI: 14 560 - 15 847), which accounted for 1.28% (15 202/1 190 340) of the total number of males aged from 15 to 49 and the ration of clients to female sex worker was 2.39:1. CONCLUSION: NSUM is an easy and quick way to estimate the size of female sex workers or clients, but the estimated sizes are subject to bias and error due to estimate effect and sample representativeness.

Adsorption characteristics of SO2 onto novel activated carbon fixed bed: kinetics, isotherms, thermodynamics and washing regeneration.

The problem of SO2 pollution in industrial flue gas has brought great pressure to environmental governance. In this study, a new type of activated carbon fixed bed device was designed and built for flue gas desulfurization. The results showed that activated carbons (AC1-AC5) were microporous activated carbons with abundant functional groups on the surface, and the desulfurization performance was ranked as AC1 > AC2 > AC3 > AC4 > AC5. The specific surface area of AC1 was as high as 624.98 m2/g, and the maximum adsorption capacity was 29.03 mg·g-1 under the optimum reaction conditions. The Freundlich adsorption isotherm model and Bangham pore diffusion model are more suitable for describing the dynamic adsorption process of SO2 on AC1. Combined with thermodynamic research, it is shown that the adsorption process of SO2 is a spontaneous, exothermic, and chaotic reduction process, which is mainly a physical adsorption between single-layer adsorption and multi-layer adsorption. Finally, the desulfurization-washing regeneration cycle experiment results showed that the regeneration rate of AC1 increases with the washing time and washing temperature, up to 95%, which provides data reference for industrial application.

Morphological component analysis under non-convex smoothing penalty framework for gearbox fault diagnosis.

The sparse representation methodology has been identified to be a promising tool for gearbox fault diagnosis. The core is how to precisely reconstruct the fault signal from noisy monitoring signals. The non-convex penalty has the ability to induce sparsity more efficiently than convex penalty. However, the introduction of non-convex penalty usually influences the convexity of the model, resulting in the unstable or sub-optimal solution. In this paper, we propose the non-convex smoothing penalty framework (NSPF) and combine it with morphological component analysis (MCA) for gearbox fault diagnosis. The proposed NSPF is a unify penalty construction framework, which contains many classical penalty while a new set of non-convex smoothing penalty functions can be generated. These non-convex penalty can guarantee the convexity of the objective function while enhancing the sparsity, thus the global optimal solution can be acquired. The simulation and engineering experiments validate that the NSPF enjoys more reconstruction precision compared to the existing penalties.

Immune Persistence and Safety After SARS-CoV-2 BNT162b1 mRNA Vaccination in Chinese Adults: A Randomized, Placebo-Controlled, Double-Blind Phase 1 Trial.

INTRODUCTION: BNT162b1 is a lipid nanoparticle-formulated, nucleoside-modified mRNA SARS-CoV-2 vaccine. Here, we report safety and immune persistence data following a primary two-dose vaccination schedule administered 21 days apart. METHODS: Immune persistence was determined at month 3 in 72 younger participants (aged 18-55 years) and at month 6 in 70 younger and 69 older participants (aged 65-85 years). RESULTS: In younger participants, neutralizing antibody (nAb) geometric mean titers (GMTs) for the 10 and 30 µg dose levels declined from 233 and 254 (21 days after dose 2) to 55 and 87 at month 3, respectively, and to 16 and 27 at month 6, respectively. In older participants, nAb GMTs declined from 80 and 160 (21 days after dose 2) to 10 and 21 at month 6. Overall, higher antibody titers were observed in younger participants, and the 30 µg dose induced higher levels of nAb, which declined more slowly by month 6. No serious adverse events were reported in the vaccine group. CONCLUSION: This study showed BNT162b1 maintains a favorable safety profile in younger and older participants in the 6 months after vaccination. This study further extends our understanding of immune persistence and the safety of the BNT162b1 vaccine as a candidate vaccine in the BioNTech pipeline. TRIAL REGISTRATION NUMBER: NCT04523571, registered August 21, 2020.

Safety and immunogenicity of the SARS-CoV-2 BNT162b1 mRNA vaccine in younger and older Chinese adults: a randomized, placebo-controlled, double-blind phase 1 study.

An effective vaccine is needed to end the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic. Here, we assess the preliminary safety, tolerability and immunogenicity data from an ongoing single-center (in Jiangsu province, China), parallel-group, double-blind phase 1 trial of the vaccine candidate BNT162b1 in 144 healthy SARS-CoV-2-naive Chinese participants. These participants are randomized 1:1:1 to receive prime and boost vaccinations of 10 µg or 30 µg BNT162b1 or placebo, given 21 d apart, with equal allocation of younger (aged 18-55 years) and older adults (aged 65-85 years) to each treatment group (ChiCTR2000034825). BNT162b1 encodes the SARS-CoV-2 spike glycoprotein receptor-binding domain (RBD) and is one of several messenger RNA-based vaccine candidates under clinical investigation. Local reactions and systemic events were generally dose dependent, transient and mild to moderate. Fever was the only grade 3 adverse event. BNT162b1 induced robust interferon-γ T cell responses to a peptide pool including the RBD in both younger and older Chinese adults, and geometric mean neutralizing titers reached 2.1-fold (for younger participants) and 1.3-fold (for the older participants) that of a panel of COVID-19 convalescent human sera obtained at least 14 d after positive SARS-CoV-2 polymerase chain reaction test. In summary, BNT162b1 has an acceptable safety profile and produces high levels of humoral and T cell responses in an Asian population.

Safety and immunogenicity of a recombinant COVID-19 vaccine (Sf9 cells) in healthy population aged 18 years or older: two single-center, randomised, double-blind, placebo-controlled, phase 1 and phase 2 trials.

COVID-19 vaccines from multiple manufacturers are needed to cope with the problem of insufficient supply. We did two single-center, randomised, double-blind, placebo-controlled phase 1 and phase 2 trials to assess the safety, tolerability and immunogenicity of a recombinant COVID-19 vaccine (Sf9 cells) in healthy population aged 18 years or older in China. Eligible participants were enrolled, the ratio of candidate vaccine and placebo within each dose group was 3:1 (phase 1) or 5:1 (phase 2). From August 28, 2020, 168 participants were sequentially enrolled and randomly assigned to receive the low dose vaccine, high dose vaccine or placebo with the schedule of 0, 28 days or 0, 14, 28 days in phase 1 trial. From November 18, 2020, 960 participants were randomly assigned to receive the low dose vaccine, high dose vaccine or placebo with the schedule of 0, 21 days or 0, 14, 28 days in phase 2 trial. The most common solicited injection site adverse reaction within 7 days in both trials was pain. The most common solicited systematic adverse reactions within 7 days were fatigue, cough, sore throat, fever and headache. ELISA antibodies and neutralising antibodies increased at 14 days, and peaked at 28 days (phase 1) or 30 days (phase 2) after the last dose vaccination. The GMTs of neutralising antibody against live SARS-CoV-2 at 28 days or 30 days after the last dose vaccination were highest in the adult high dose group (0, 14, 28 days), with 102.9 (95% CI 61.9-171.2) and 102.6 (95% CI 75.2-140.1) in phase 1 and phase 2 trials, respectively. Specific T-cell response peaked at 14 days after the last dose vaccination in phase 1 trial. This vaccine is safe, and induced significant immune responses after three doses of vaccination.

Non-dominated solution set based on time-frequency infograms for local damage detection of rotating machines.

The determination of an index to balance the impulsiveness and cyclostationarity of an expected component is an interesting research topic in mechanical health monitoring. Many proposed indices focus on evaluating the impulsiveness or cyclostationarity of an analyzed signal, thus resulting that an expected component cannot dominate other modes using these indices under the interferential circumstance of occasional shocks and harmonic signals. Moreover, several parameters have to be determined in advance, thereby leading to taxing consumption even if some methods can achieve a trade-off between impulsiveness and cyclostationarity via multi-objective optimization algorithms. A non-dominated solution set based on time-frequency infograms is then proposed in this study to preferably synthesize the impulsiveness and cyclostationarity of detecting the local damage of rotating machines. Specifically, the proposed method is conducted through four steps: (a) decompose the measured signal into certain levels, (b) construct the time-frequency infograms, (c) define the non-dominated solution set by Pareto filtering, and (d) perform Hilbert transform on these non-dominated solutions to diagnose the health condition of the rotating machines. Several case studies, including a simulated signal, and experimental data from defective bearing and gear respectively, were conducted to validate the proposed method. Analysis and comparison results indicated that the proposed method outperforms the fast kurtogram, infograms and some advanced indices in identifying fault features from rotating machines.

[Kinetics model of spherical immobilized cellulase].

A kinetics model was developed for predicting and simulating immobilized cellulase performance, which follows Michaelis-Menten kinetics with competitive product inhibition. Taking into account the effects of competitive product inhibition, inner diffusional limitation, substrate concentration and carrier size, the substrate distribution and the product distribution in carriers were investigated, and the effectiveness factors were also calculated over a wide range of parameters. The effects of competitive product inhibition are shown to increase the substrate concentration in the carrier, and, additionally, to increase the effectiveness factors slightly. With the increase of inner diffusion coefficient, both the effectiveness factors and the substrate concentration in the carrier increase. As the carrier size increases, on the other hand, these values decrease. The effectiveness factors and the substrate concentration in the carrier are found to increase when substrate concentration in the reaction system increases.