Deep learning (DL) has become a popular option for data-driven fault diagnosis, because it can avert the influence of subjective factors in an artificial feature extraction process. However, it also suffers from the adverse effects accompanied with small fault sample and unbalanced data, resulting in limited accuracy improvement. For the aforementioned problem, this paper introduces a variational auto-encoder (VAE) into a fault diagnosis framework to realize data amplification by vibration signal generation, then an enhanced fault diagnosis approach is proposed combining with a convolution neural network. The well-trained VAE can realize the infinite generation of vibration signal by using the hidden variables sampled from Gaussian distribution, and then the generated artificial signals are mixed with the real original signals to form an enhanced training set, which can be utilized for classifier training to realize fault identification. Experimental results show that the generated artificial signals have similar time-frequency characteristics compared with the original real ones, and the enhanced fault diagnosis method holds a higher and more stable recognition accuracy than the unenhanced version and other typical methods.

Polypropylene/high-density polyethylene (PP/HDPE) blends and their nanocomposites are melt compounded using twin-screw extruder. The as-extruded samples are then batch foamed with supercritical carbon dioxide. It is demonstrated that the blend with a viscosity ratio close to unity produces a microcellular foam with very uniformly distributed finer cell structure, whereas nonuniformly distributed cells may be obtained in the foamed blend with a viscosity ratio greater than unity. Then three methods are tried to improve the cellular structure of the latter. The results show that the increase of HDPE content improves the uniformity of cellular size and distribution, but results in the increase of cell diameter and the decrease of cell density. Properly lowering the foaming temperature reduces the cell diameter and improves the cellular distribution. Incorporating nanoparticles, including nano-calcium carbonate and nanoclay, into the blend leads to a significantly smaller cell size and more homogeneous cell distribution. A critical content of nanoparticles exists to obtain dramatically improved cell structure. Moreover, the results show that most cells in foamed blends are nearly reticulated. So open-celled foams are achieved in this work.

No abstract

An on-chip refractive index (RI) sensor, which is based on the localized surface plasmon resonance (LSPR) of periodic gold nanorings array, is presented. The structure parameters and performance of LSPR-based sensors are optimized by analyzing and comparing the LSPR extinction spectra. The mechanism of the enhancement of plasma resonance in a ring array is discussed by the simulation results. A feasible preparation scheme of the nanorings array is proposed and verified by coating a gold film and etching on the photonic crystals. Based on the optimum sensing structure, an RI sensor is constructed with a RI sensitivity of 577 nm/refractive index unit (RIU) and a figure of merit (FOM) of 6.1, which is approximately 2 times that of previous reports.

Secretory immunoglobulin A (SIgA) is a non-inflammatory antibody that shields internal body surfaces, such as in the intestine to neutralize pathogens in the lumen of the intestine. As chemotherapy seriously damages the mucosal immune system, we herein demonstrated that polysaccharide from the squid ink of Ommastrephes bartrami (OBP) activated intestinal SIgA secretion to prevent chemotherapeutic injury. Using a mouse model of chemotherapy induced intestinal injury by intraperitoneal injection of 50 mg kg(-1) cyclophosphamide, our results showed an enhanced SIgA concentration in intestinal mucosa by OBP administration and the higher production of SIgA relied on the greater expression of IgA, J chain and pIgR. Furthermore, the higher expressions of IL-6, IL-10 and TNF-α increased by OBP treatment contributed to enhanced IgA and J chain synthesis in IgA(+) plasma cells, and pIgR expression in epithelial cells. It also triggered a prompt immunoglobulin secretory pathway confirmed by enhanced UPR (unfolded protein response) effectors XBP-1s and Bip expression. Our results have important implications for the mucosal immunity enhancement effects of OBP as a functional food component for chemotherapeutic patients.

No abstract

We study the exotic deformed nucleus 31 Ne using an approach that combines self-consistent structure and reaction theory. We utilize the fully-relativistic, microscopic deformed Hartree-Bogoliubov theory in continuum (DRHBc) to demonstrate that deformation and pairing correlations give rise to a halo structure with large-amplitude p-wave configuration in 31 Ne. We then use the valence nucleon wave functions and angle-averaged density distributions of 30 Ne from this model as input for a Glauber reaction model to study the observables of neutron-rich Neon isotopes and search for halo signatures. Our predictions of the reaction cross sections of these exotic Neon isotopes on a Carbon target can better reproduce the experimental data than those from relativistic mean field model for a spherical shape with resonances and pairing correlations contributions, as well as those from a Skyrme-Hartree-Fock model. The one-neutron removal cross section at 240 MeV/nucleon, the inclusive longitudinal momentum distribution of the 30 Ne, and the valence neutron residues from the 31 Ne breakup reaction are largely improved over previous theoretical predictions and agree well with data. These reaction data indicate a dilute density distribution in coordinate space and are a canonical signature of a halo structure.

No abstract