No abstract

One of the main challenges that the industry faces when dealing with massive data for failure diagnosis is high dimensionality of such data. This can be tackled by dimensionality reduction method such as principal components analysis, which usually results in an improved fault diagnosis. Other available techniques include auto-encoders and its variants denoising auto-encoders and sparse auto-encoders. Most recently, variational auto-encoders are one of the most promising techniques for unsupervised learning with successful applications in image processing and speech recognition. Differently from other auto-encoder methods, variational auto-encoders use variational inference to generate a latent representation of the data and impose a distribution over the latent variables and the data itself. In this article, we propose a fully unsupervised deep variational auto-encoder-based approach for dimensionality reduction in fault diagnosis and explore the variational auto-encoder capabilities for such a task. This is achieved by comparing the latent representations provided by variational auto-encoders to the ones from principal components analysis as well as when no reduction is performed in ball bearings' fault classification using vibration signals. To tackle massive sensor data, we propose different architectures for the variational auto-encoder's encoder and decoder that are based on deep neural networks and deep convolutional neural networks. Experiments are also carried out by varying the data preprocessing methods for generating spectrograms and hand-engineering features as well as the use of raw vibration data, the architecture of the neural networks fault classifiers operating on the latent representations from variational auto-encoder and principal components analysis, the degree of data dimension reduction, and the size of the available labeled data used for training the fault classifiers. The results show that variational auto-encoders are a competent and promising tool for dimensionality reduction for use in fault diagnosis and worth further exploring their capabilities beyond vibration signals of ball bearing elements.

BackgroundThe extract from Moringa oleifera seeds is used worldwide, especially in rural areas of developing countries, to treat drinking water. M. oleifera seeds contain the lectins cmol and WSMoL, which are carbohydrate-binding proteins that are able to reduce water turbidity because of their coagulant activity. Studies investigating the ability of natural products to damage normal cells are essential for the safe use of these substances. This study evaluated the cytotoxic and anti-inflammatory properties of the aqueous seed extract, the extract used by population to treat water (named diluted seed extract in this work), and the isolated lectins cmol and WSMoL. Methodology/Principal FindingsThe data showed that the aqueous seed extract and cmol were potentially cytotoxic to human peripheral blood mononuclear cells, while WSMoL and diluted seed extract were not cytotoxic. The M. oleifera aqueous seed extract and the lectins cmol and WSMoL were weakly/moderately cytotoxic to the NCI-H292, HT-29 and HEp-2 cancer cell lines and were not hemolytic to murine erythrocytes. Evaluation of acute toxicity in mice revealed that the aqueous seed extract (2.000 mg/kg) did not cause systemic toxicity. The aqueous seed extract, cmol and WSMoL (6.25 µg/mL) and diluted seed extract at 50 µg/mL exhibited anti-inflammatory activity on lipopolyssaccharide-stimulated murine macrophages by regulating the production of nitric oxide, TNF-α and IL-1β. The aqueous seed extract reduced leukocyte migration in a mouse model of carrageenan-induced pleurisy; the myeloperoxidase activity and nitric oxide, TNF-α and IL-1β levels were similarly reduced. Histological analysis of the lungs showed that the extract reduced the number of leukocytes. Conclusion/SignificanceThis study shows that the extract prepared according to folk use and WSMoL may be non-toxic to mammalian cells; however, the aqueous seed extract and cmol may be cytotoxic to immune cells which may explain the immunosuppressive potential of the extract.

Summary The resistance of micro‐organisms to antimicrobial agents has been a challenge to treat animal and human infections, and for environmental control. Lectins are natural proteins and some are potent antimicrobials through binding to carbohydrates on microbial surfaces. Oligomerization state of lectins can influence their biological activity and maximum binding capacity; the association among lectin polypeptide chains can alter the carbohydrate–lectin binding dissociation rate constants. Antimicrobial mechanisms of lectins include the pore formation ability, followed by changes in the cell permeability and latter, indicates interactions with the bacterial cell wall components. In addition, the antifungal activity of lectins is associated with the chitin‐binding property, resulting in the disintegration of the cell wall or the arrest of de novo synthesis from the cell wall during fungal development or division. Quorum sensing is a cell‐to‐cell communication process that allows interspecies and interkingdom signalling which coordinate virulence genes; antiquorum‐sensing therapies are described for animal and plant lectins. This review article, among other approaches, evaluates lectins as antimicrobials.

Moringa oleifera flower extract, with trypsin inhibitor activity, is a larvicidal agent on Aedes aegypti. This work reports the isolation of trypsin inhibitor (M. oleifera flower trypsin inhibitor (MoFTI)) and its effect on A. aegypti egg hatching, viability of newly hatched larvae, survival of pupae, and growth of inhabitant bacteria from midgut of fourth-instar larvae (L4). MoFTI (K i, 2.4 μM), isolated by affinity chromatography on trypsin-agarose column, was an 18.2 kDa polypeptide on sodium dodecyl sulphate-polyacrylamide gel electrophoresis. Flower extract (at concentrations of 8.5-17.0 mg/mL) reduced egg hatchability while MoFTI (0.05-0.5 mg/mL) did not affect the hatching rate. Mortality of newly hatched larvae ranged from 3.5 to 19.1 % in the presence of the extract (4.0-17.0 mg/mL) and was also promoted by MoFTI (LC50, 0.3 mg/mL). After 72 h, larvae incubated with extract at 13.0 and 17.0 mg/mL were at stages L2 and L1, respectively, while in control they reached L3 instar. In the presence of MoFTI, at all concentrations tested, the larvae did not pass the first instar. Flower extract and MoFTI did not interfere on pupae survival. The extract and MoFTI inhibited the growth of L4 gut bacteria (minimum inhibitory concentrations of 3.47 and 0.031 mg/mL, respectively) but only the inhibitor showed bactericide effect (minimum bactericidal concentration of 1.0 mg/mL). The findings reported herein indicate that MoFTI constitutes a larvicidal principle from M. oleifera flowers against A. aegypti newly hatched larvae and is an antibacterial agent active against the microbiota from L4 gut.