Intelligent Fault Diagnosis Method of Rolling Bearings Based on Transfer Residual Swin Transformer with Shifted Windows

Due to their robust learning and expression ability for complex features,the deep learning(DL)model plays a vital role in bearing fault diagnosis.However,since there are fewer labeled samples in fault diagnosis,the depth of DL models in fault diagnosis is generally shallower than that of DL models in other fields,which limits the diagnostic performance.To solve this problem,a novel transfer residual Swin Transformer(RST)is proposed for rolling bearings in this paper.RST has 24 residual self-attention layers,which use the hierarchical design and the shifted window-based residual self-attention.Combined with transfer learning techniques,the transfer RST model uses pre-trained parameters from ImageNet.A new end-to-end method for fault diagnosis based on deep transfer RST is proposed.Firstly,wavelet transform transforms the vibration signal into a wavelet time-frequency diagram.The signal’s time-frequency domain representation can be represented simultaneously.Secondly,the wavelet time-frequency diagram is the input of the RST model to obtain the fault type.Finally,our method is verified on public and self-built datasets.Experimental results show the superior performance of our method by comparing it with a shallow neural network.

SAR Azimuth Cut-off to Estimate Wind Speed under High Wind Regimes

In this study,the azimuth cut-off method,typically used for SAR moderate wind speed estimation purposes,is analyzed under high wind regimes.Firstly,the importance of the pixel spacing,the size of the boxes selected for Synthetic Aperture Radar(SAR)image partitioning and the image texture in terms of homogeneities are discussed by considering their influence on the azimuth cut-off(λc)estimation.Secondly,a quality control analysis of the reliability ofλc is carried out by evaluating the distance between the autocorrelation functions(ACF)and their correspondent fittings.This analysis points out the importance of filtering out the unreliable and unfeasibleλc values in order to improve the wind speed estimation.The quality control procedure is based on a x2 test,applied on a large Sentinel-1 A dataset.The soundness of the test is verified by an increment in terms of correlation betweenλc estimations and wind speed values.This approach is,then,applied under high wind regimes,i.e.,tropical cyclones.

Modeling projected changes of mangrove biomass in different climatic scenarios in the Sunda Banda Seascapes

Mangroves are critical in the ecological,economic and social development of coastal rural and urban communities.However,they are under threat by climate change and anthropogenic activities.The Sunda Banda Seascape(SBS),Indonesia,is among the world’s richest regions of mangrove biomass and biodiversity.To inform current and future management strategies,it is critical to provide estimates of how mangroves will respond to climate change in this region.Therefore,this paper utilized spatial analysis with model-based climatic indicators(temperature and precipitation)and mangrove distribution maps to estimate a benchmark for the mangrove biomass of the SBS in six scenarios,namely the Last Inter-glacial Period,the current scenario(1950-2000)and all four projected Representative Concentration Pathways in 2070 due to climate change.Despite mangroves gaining more biomass with climate change(the increase in CO_(2)concentration),this paper highlighted the great proportion of below-ground biomass in mangrove forests.It also showed that the changes in spatial distribution of mangrove biomass became more variable in the context of climate change.As mangroves have been proposed as an essential component of climate change strategies,this study can serve as a baseline for future studies and resource management strategies.

Chemically Modified Uridine Molecules Incorporating Acyl Residues to Enhance Antibacterial and Cytotoxic Activities

A new N-acetylsulfanilylation series of uridine have been synthesized in good yield using direct acylation method and afforded the 5’-O-N-acetylsulfanilyluridine. In order to obtain newer products, the 5’-O-N-acetylsulfanilyluridine derivative was further transformed to a series of 2’,3’-di-O-acyl derivatives containing a wide variety of functionalities in a single molecular framework. The chemical structures of the newly synthesized compounds were confirmed on the basis of their FTIR, 1H-NMR spectroscopy, physicochemical properties and elemental analysis. All the synthesized uridine derivatives were tested for their in vitro antibacterial activity against six human pathogenic bacterial strains and for comparison standard antibiotic Ampicillin was also determined. The study revealed that the selectively acylated deriva-tives 5’-O-N-acetylsulfanilyl-2’,3’-di-O-lauroyluridine and 5’-O-N-acetylsulfanilyl-2’,3’-di-O-pivaloyluridine showed highest inhibition against Staphylococcus aureus and Bacillus cereus, respectively. We also observed that the introduction of hexanoyl, decanoyl, lauroyl, myristoyl and pivaloyl groups, the antibacterial functionality of the compound uridine increases. Another noteworthy observation was that the uridine derivatives were found comparatively more effective against Gram-positive microorganisms than those of Gram-negative microorganisms. In addition, the test chemicals were also tested for cyto-toxicity by brine shrimp lethality bioassay and compounds showed different rate mortality with different concentrations.