Search

Recent progress on intelligent fault diagnosis is mainly attributed to the explosive development of convolutional neural networks (CNN). Many existing CNN-based fault diagnosis models can extract abundant features from the measured vibration signals but cannot explore enough discriminative features under strong noise conditions. This poses a challenge for industrial applications. To address this problem, we develop a new deep CNN model, called a multireceptive field denoising residual convolutional network (MF-DRCN). The major contributions are: 1) a multireceptive field denoising (MFD) block is designed to enhance the deep features extracted by the CNN model and filter out the interference feature information; 2) an adaptive feature integration (AFI) module is embedded in the CNN model to adaptively integrate features, so as to make better use of the extracted information; and 3) an end-to-end CNN model called MF-DRCN is developed based on MFD and AFI. The experimental results demonstrate that MF-DRCN has better feature extraction and anti-interference capabilities than the other seven competitive methods. Specifically, under strong noise conditions with SNR = -6 dB, MF-DRCN achieves 84.51% and 86.45% diagnostic accuracy respectively on the planetary gearbox data set and the industrial pump data set, which suggests MF-DRCN is a promising intelligent fault diagnosis approach.

In this paper, the Plateau–Rayleigh (PR) instability that occurs in the two-phase fluids is numerically investigated with dissipative particle dynamics (DPD) method at the mesoscale particle level. For modeling two-phase fluid, the “color” repulsion model is applied to depict binary fluids according to Rothman–Keller method. The present DPD simulation can reproduce the dynamics behaviors of PR instability. Moreover, we research the influence of the ratio of width and length of the liquid column on the PR instability. The results show that the number of droplets formed due to the PR instability increases as the ratio decreases, which is linear relationship within a certain range. Furthermore, some small-scale characteristics, which are difficult to be captured in macroscopic numerical simulation, can be observed by the DPD simulation.

The study of the effect of pressure and temperature on the structural and mechanical properties of materials is very difficult for current experimental techniques. Here, we have synthesized the γ-Nb5Si3 compound with remarkable quality by high pressure and high temperature (HPHT) technology. The effects of high pressure and sintering temperature on its microstructure, mechanical properties, and oxidation resistance are discussed. The Vickers hardness and density of the γ-Nb5Si3 sample at 5.5 GPa and 1100 °C are 8.67 GPa and 5.73 g/cm3, respectively. It is shown that the HPHT behavior not only improves sample's compactness and microstructure, but also makes γ-Nb5Si3 to have a strong response to the Vickers hardness and related density. These results effectively demonstrate the functions of the HPHT technology, which accurately express the Vickers hardness, density, and thermal stability of γ-Nb5Si3 compounds.

The effect of germanium content on the structural, electronic, and optical properties of ordered Si 1 − x Ge x C alloys ( x = 0, 0.25, 0.5, 0.75, 1) has been systemically investigated in the present work by using first principles calculations based on the density functional theory. The calculated lattice parameters, bulk and shear moduli which exhibit linear dependences on the composition x follow the Vegard’s law and are fitted by linear interpolation. The electronic band structure calculations performed by the application of HSE03 functional reveal that Si 1 − x Ge x C alloys are all wide band gap semiconductors, among which Si 0.75 Ge 0.25 C and Si 0.25 Ge 0.75 C have direct band gaps. The substitution of Ge for Si changes the symmetry of zb-SiC unit cell, giving rise to the indirect to direct band gap transition. The composition dependent static optical functions, such as dielectric constants, refractive index and Plasmon energy are discussed.

First principles calculations were performed to study the mechanism of nitric oxide release through the catalytic decomposition of S-nitrosothiols mediated by gold nanoparticles within the density functional theory. The gold nanoparticles and S-nitrosothiols were modeled by Au38 cluster and SPEN compound, respectively. The results reveals a two-step reaction process. The primary step is SPEN adsorbs on Au38 cluster surface by S atom to form the intermediate. In the next step, NO dissociates from SPEN under the catalysis of Au38 cluster and thereafter a specie of Au-thiolate complex was produced as the product. Furthermore, we found that the overall catalytic reaction is exothermic with a small energy barrier, which confirms the reaction process is kinetically feasible.

A systematic theory of holographic optical disk storage is developed. The recorded modes of addressing/tracking information on a holographic optical disk are discussed in detail, then the data storage format of a holographic optical disk is proposed. Parallel addressing techniques of the recorded images are analyzed.

A theoretical prediction on the structural stabilities, mechanical properties, and electronic properties of the C-centered orthorhombic (Cco) Si(8) and Ge(8) is presented, inspired by a recently proposed carbon allotrope structure, Cco-C(8). Energetically comparable with previously known metastable phases, Cco-Si(8) and Cco-Ge(8) may be obtained by decompressing the high-pressure β-Sn phases, or by compressing the corresponding nanotubes. The calculated bulk moduli of Cco-Si(8) and Cco-Ge(8) are close to those of the diamond phases. Further study of the electronic properties reveals that the band gaps of Cco-Si(8) and Cco-Ge(8) are tunable with variations in lattice parameters.

The room-temperature structural phase transition of graphite at elevated hydrostatic pressure has drawn considerable scientific interest for its fundamental importance in condensed matter physics and materials science over the past few decades. A pressure-induced phase transition has been demonstrated in previous experiments by the measurement of electrical resistance, optical spectrum, X-ray diffraction spectrum, inelastic X-ray scattering and Raman spectroscopy. However, the nature of the cold-compressed graphite phase has been puzzling the experts and pioneers in the field of high pressure research due to some inherent factors, until recently a monoclinic structure, i.e. [Formula: see text]-carbon, stands out from the other structure candidates and successfully accounts for the crystal structure of the cold-compressed graphite phase both in theory and experiment that eventually putting an end to this long-lasting controversial issue. This paper reviews the recent progress on the pressure-induced phase transitions of graphite at room temperature especially for the theoretical investigations. The review will focus on the recent proposed novel carbon allotropes as candidate structures of the cold-compressed graphite phase by using different crystal structure prediction methods. The history of structure determination of cold-compressed graphite phase is discussed.

The electronic and optical properties of zinc-blende (zb)[Formula: see text]GeC have been investigated using first principles calculations based on the density functional theory (DFT). The obtained band gap of zb–GeC is 2.30[Formula: see text]eV by means of Heyd–Scuseria–Ernzerhof (HSE) functional. We have discussed the energy-dependent optical functions including dielectric constants, refractive index, absorption, reflectivity, and energy-loss spectrum in detail. The results reveal that zb–GeC has a higher static dielectric constant compared with that of zb–SiC. The optical functions are mainly associated with the interband transitions from the occupied valence bands (VBs) Ge[Formula: see text][Formula: see text] and C[Formula: see text][Formula: see text] states to Ge[Formula: see text][Formula: see text], [Formula: see text] and C[Formula: see text][Formula: see text] states of the unoccupied conduction bands (CBs).

We have made significant progress towards achieving 100 Gb/in/sup 2/. We have designed and fabricated a super-solid immersion lens (SIL), implemented a new spot size measurement technique capable of measuring very small near-field spots, fabricated centre aperture detection magnetic super resolution (CAD-MSR) media and demonstrated turbo coding for ultra high recording densities. We will report the details of this effort and the recording densities that have been achieved.

The Electrooptic Beam Scanner is a solid state device developed at Carnegie Mellon's Data Storage Systems Center (DSSC) that is capable of scanning a laser beam approximately +/- 1 - 3 degree(s) at frequencies of at least 200 kHz. The optical group of the DSSC is currently employing the scanner as a fine actuator for high performance optical data storage systems. This paper discusses several aspects of that research, including multi-track scanning for increased data rate, controller for two stage actuator systems, and feedforward control for ultra-fast seeks.

Electro-optic (EO) scanners can be used as fine tracking actuators to improve the servo bandwidth in future high density/high data-rate optical disk drives. In this paper we report on the use of an EO scanner in a new optical tracking system. Track following has been accomplished with a servo bandwidth of 200 kHz and we have demonstrated track switching between nine tracks using only an EO scanner. A fine tracking experiment using an EO scanner has been demonstrated in parallel with a voice-coil actuated lens to expand the fine tracking range. Significant improvement in track switching speed and track following are demonstrated with the scanner/lens actuators as compared to tracking with the lens actuator alone.© (1999) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Optical disks are widely used in data storage system as well as in audio disk and video disk systems. Rapid progress is now being achieved in toms of recording density and data capacity. Ultra-high density magneto-optical recording has been obtained using recently developed magnetic field modulation (MFM) recording and magnetic super resolution (MSR) and magnetic domain expansion readout techniques.

High performance phase screens are essential for the focal- plane irradiance profile control in laser-fusion system. A new global/local united search algorithm is presented for constructing continuous phase screens to eliminate the wide- angle scattering. The focal-plane intensity profile fits a 16th-order flat-top super-Gaussian function. More than 98.5% of the incident energy is homogeneously constrained in the desired region. The wide-angle scattering outside (phi) 1 mm in focal plane of focus spots is less than 0.5% of the incident energy.

The mathematics relations of the system characteristics versus the material properties and read/write conditions areobtained by analysis the frequency-domain optical storage system. Through computer simulation analyses, the optimalparameter space is obtained. In addition, absorption area of material, quantum efficiency and recording light intensityare analyzed quantitatively.Keywords: frequency-domain optical storage, spectral holes, storage density, parameter space 1. INTRODUCTION When laser incidents upon the crystal material, the material will react photochemically and a series of spectral holeswill be formed. Multiple data can be recorded at the same recorded point by using different laser frequencies. This newstorage scheme has been called "frequency-domain optical storage"(FDOS), and it can increase the density of datastorage as much as iOn. The crux to increase the storage density is to form the narrow spectral holes, therefore, therequirements of material characteristics are more critical. However, the materials reported by now cannot satisfy therequirements.

A novel method for implementing joint transform correlation is presented, based on the theory of holographic optical disk storage. The specified optical disk holograms are designed by our developed iterative search algorithm. The holographic optical disk joint transform correlator and the track tracing techniques are described. The experiment results are given to demonstrate the proposed systems.© (1996) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Based on the analyses of holographic optical disk storage theory and the technological issues, the parallel addressing/tracing technology are discussed in detail. The optical comparison parallel addressing and tracing servo/track-search systems are described and experimentally demonstrated.© (1996) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Differing from the storage of image on the optical disk in series, a new method for realizing image parallelstorage is presented, based on the theory of holographic optical disk storage. The image parallel readout architecture is described, and the specified optical disk holograms are designed by our proposed iterative search algorithm. The experiment results have demonstrated the image parallel storage technique.Keywords: image parallel storage, holographic optical disk storage, optical disk holograms, image storage format,iterative search algorithm, 1. INTRODUCTION Optical disk systems have been put to practical use as large capacity storage systems. The major disadvantage ofoptical disk is the longer access time which results in lower sustained transfer rates, and the data information is retrievedby sequential reading from optical disk using a laser spot of lirn size. However, a higher data transfer rate and parallel access performance are required for many applications, especially in optical information processing and imagestorage.'31

A novel programmable optical interconnect system based on holographic optical memory disk is pre-sented to solve the contradiction of high diffractive performance and programmability in the free-spaceholographic interconnections . The parallel recording system of holographic optical memory disk is de-signed. The main advantages of the proposed system are high speed random accessibility , high storage ca-pacity and high diffractive performance. Keywords : Free-space holographic interconnections , Programmable interconnections , Holographic op- tical memory disk , Parallel recording. 1. INTRODUCTION The advantages of optical interconnections over electrical interconnections are high speed massive par- allelism , and lower mutual interference'. Free- space interconnection is the most attractive among the oth-ers for its potential massive parallelism. Both classical and holographic optical elements are being consid-ered candidates for free-space interconnection , however , holographic

In this paper we report a method for producing very small aperture lasers (VSALs) from low-cost, commercial index guided edge emitting diode lasers. The detailed fabrication procedure is provided and the VSALs are characterized. Both high power and high aperture power to background noise power ratio are obtained even with a small aperture. This fabrication method can be applied to most commercial diode lasers without knowledge of the detailed geometry and is highly reproducible.

Beam deflectors can be used as fine tracking actuators to improve track access time and data rate in future high performance optical disk drives. In this paper we report on the use of an electro-optic (EO) scanner for optical data storage. Track following has been accomplished using this EO actuator with a servo bandwidth of 200 kHz, and single-stage high-speed track switching/following has been demonstrated in a new optical head tracking system with reduced offset. A fine tracking experiment has also been demonstrated using an EO actuator with a voice coil motor (VCM) actuator to extend the fine tracking range. A new compensator design method, the PQ method, has been used for this scanner/VCM compound actuator system. Significant improvements in track switching/following speed are demonstrated with the scanner/VCM compound actuator as compared to tracking with the VCM actuator alone.

A novel holographic frequency-domain optical storage technique is investigated. The mathematics relations of the system characteristics versus material properties and read/write conditions are developed. The system performances: recording density, signal-noise-ratio, parallel access, are analyzed in detail. The results show that this technique can realize high performance storage.