An SVM—ANN Hybrid Classifier for Diagnosis of Gear Fault

A hybrid classifier obtained by hybridizing Support Vector Machines (SVM) and Artificial Neural Network (ANN) classifiers is presented here for diagnosis of gear faults. The distinctive features obtained from vibration signals of a running gearbox, which was operated in normal and fault-induced conditions, were used to feed the SVM-ANN hybrid classifier. Time-domain vibration signals were divided in segments. Features such as peaks in time domain and in spectrum, central moments, and standard deviations were obtained from signal segments. Based on the experimental results, it was shown that SVM-ANN hybrid classifier can successfully identify gear condition and that the hybrid SVM-ANN classifier performs much better than standard versions of ANNs and SVM. The effectiveness of the hybrid classifier under noise was also investigated. It was shown that if vibration signals are preprocessed by Discrete Wavelet Transform (DWT), efficacy of the SVM-ANN hybrid is significantly enhanced.     

Effect of annealing on magnetic properties and silicide formation at Co/Si interface

The interaction of Co (30 nm) thin films on Si (100) substrate in UHV using solid state mixing technique has been studied. Cobalt was deposited on silicon substrate using electron beam evaporation at a vacuum of 4×10^?8 Torr having a deposition rate of about 0·1 Å/s. Reactivity at Co/Si interface is important for the understanding of silicide formation in thin film system. In the present paper, cobalt silicide films were characterized by atomic force microscopy (AFM) and secondary ion mass spectroscopy (SIMS) in terms of the surface and interface morphologies and depth profile, respectively. The roughness of the samples was found to increase up to temperature, 300°C and then decreased with further rise in temperature, which was due to the formation of crystalline CoSi₂ phase. The effect of mixing on magnetic properties such as coercivity, remanence etc at interface has been studied using magneto optic Kerr effect (MOKE) techniques at different temperatures. The value of coercivity of pristine sample and 300°C annealed sample was found to be 66 Oe and 40 Oe, respectively, while at high temperature i.e. 748°C, the hysteresis disappears which indicates the formation of CoSi₂ compound.     

Deposition of hBN+Cu layer through electrical discharge process using green compact electrode

In the current investigation, solid-lubricant surfaces were prepared onto the mild steel work piece by means of electrical discharge coating process in electrical discharge machine (EDM). Hexagonal boron nitride (hBN) and copper (Cu) powder-based green compact electrodes, prepared in hot mounting press were used as an EDM tool. The loosely bonded material particles from the compact tool were transported to the mild steel surface. With the change of process parameters (powder mixture ratio, duty factor, and peak current), the variation in surface morphology, micro-hardness and tribological properties of hBN-Cu coating layer has been discussed. Following the investigation, field emission scanning electron microscopic (FESEM) images of coating indicated fewer gaps and pores for the powder lubricant mixing ratio of hBN: Cu/50:50, peak current value of 10 A and duty factor value of 70%. X-ray diffraction plot composed of prominent peaks of BN, Cu, along with Fe₂O₃ (intermetallic compound). Micro-hardness value of lubricant surfaces drastically reduced to the minimum of 75.76 HV (coating layer) from 180 HV (substrate). Moreover, pin on disk tribological test indicated the lubrication effect of hBN particles by revealing a major drop in wear value from 95.75 µm (work piece) to 1.52 µm (coating) and friction coefficient from 0.9 (work piece) to 0.1 (coating).     

Discrete Fractional‐Order BAM Neural Networks with Leakage Delay: Existence and Stability Results

In this paper, a discrete model of fractional‐order bidirectional associative memory (BAM) neural networks with leakage delay is considered. Sufficient conditions are established for the existence, uniqueness and uniform stability of nontrivial solutions of the addressed model. Our approach is based on the implementation of the newly defined discrete fractional calculus, mathematical inequalities and fixed point theorems. We provide an example that supports the application of the theoretical results.     

Boosting the Efficiency of SnO2‐Triple Cation Perovskite System Beyond 20% Using Nonhalogenated Antisolvent

Solution‐processed triple‐cation perovskite solar cells (PSCs) rely on complex compositional engineering or delicate interfacial passivation to balance the trade‐off between cell efficiency and long‐term stability. Herein, the facile fabrication of highly efficient, stable, and hysteresis‐free tin oxide (SnO₂)‐based PSCs is demonstrated with a champion cell efficiency of 20.06% using a green, halogen‐free antisolvent. The antisolvent, composed of ethyl acetate (EA) solvent and hexane (Hex) in different proportions, works exquisitely in regulating perovskite crystal growth and passivating grain boundaries, leading to the formation of a crack‐free perovskite film with enlarged grain size. The high quality perovskite film inhibits carrier recombination and substantially improves the cell efficiency, without requiring an additional enhancer/passivation layer. Furthermore, these PSCs also demonstrate remarkable long‐term stability, whereby unencapsulated cells exhibit a power conversion efficiency (PCE) retention of ≈71% after >1500 hours of storage under ambient condition. For encapsulated cells, an astounding PCE retention of >98% is recorded after >3000 hours of storage in air. Overall, this work realizes the fabrication of SnO₂‐based PSCs with a performance greater or comparable to the state‐of‐the‐art PSCs produced with halogenated antisolvents. Evidently, EA–Hex antisolvent can be an extraordinary halogen‐free alternative in maximizing the performance of PSCs.     

Subspace histograms for outlier detection in linear time

Outlier detection algorithms are often computationally intensive because of their need to score each point in the data. Even simple distance-based algorithms have quadratic complexity. High-dimensional outlier detection algorithms such as subspace methods are often even more computationally intensive because of their need to explore different subspaces of the data. In this paper, we propose an exceedingly simple subspace outlier detection algorithm, which can be implemented in a few lines of code, and whose complexity is linear in the size of the data set and the space requirement is constant. We show that this outlier detection algorithm is much faster than both conventional and high-dimensional algorithms and also provides more accurate results. The approach uses randomized hashing to score data points and has a neat subspace interpretation. We provide a visual representation of this interpretability in terms of outlier sensitivity histograms. Furthermore, the approach can be easily generalized to data streams, where it provides an efficient approach to discover outliers in real time. We present experimental results showing the effectiveness of the approach over other state-of-the-art methods.     

POSSIBILITY OF CONCENTRATION POLARIZATION OCCURRING INSIDE THE MEMBRANE DURING STEADY STATE PERVAPORATION

Mathematical equations are derived for describing the pervaporation transport of pure penetrant through polymeric membranes by assuming the chemical potential gradient as the driving force for the flow of penetrant. An imaginary phase (liquid or vapor) is assumed to be present and is in thermodynamic equilibrium with the membrane phase for deriving these equations. The mathematical equations obtained are similar to those derived by Okada and Matsuura (1991) using the pore-flow model. The possibility of concentration polarization occurring inside the membrane is predicted based on the analysis of binary mixture pervaporation. Experimental profiles of binary penetrant concentration in the membrane are established and these profiles substantiate the prediction of concentration polarization occurring inside the membrane. Pervaporation and sorption data from liquid and vapor phase at 25° C are reported for the acetic acid-water-polyamide system.     

Inverse gas chromatography of poly(n-butyl methacrylate): Effect of flow rate on specific retention volume and detection of glass transition temperature

Flow-rate effect on specific retention volume (V) was studied by eluting aliphatic, aromatic, and chlorinated aliphatic probes at infinite dilution on poly(n-butyl methacrylate) stationary phase at different temperatures from ?10 to 150°C, encompassing both the glass transition (T_g) and the softening temperatures of the polymer. The effect became pronounced as the temperature was reduced below 100°C. V decreased with an increase in the flow rate: first linearly at temperatures between 70 and 100°C, and then nonlinearly at all temperatures below 70°C. The retention diagrams of n-pentane, isooctane, and cyclohexane alone enabled the detection of glass transition. Dichloromethane gave a linear retention diagram through T_g without showing the flow-rate effect.     

Utility of PCR assay in diagnosis of en-plaque tuberculoma of the brain

This is the first report of a case in which diagnosis of en-plaque tuberculoma on the basis of magnetic resonance imaging (MRI) findings was confirmed by a Mycobacterium tuberculosis complex-specific PCR assay of cerebrospinal fluid. The accuracy of the diagnosis was supported by good response to antitubercular drugs, which was shown by repeat MRI studies performed after treatment.     

Preparation and bulk thermal expansion studies in M1−xCexSiO4 (M = Th, Zr) system, and stabilization of tetragonal ThSiO4

Two series of compositions with the general formula M_1−xCe_xSiO₄ (M = Th, Zr; x = 0.0–0.5; 1.0) were prepared by a standard solid state route and characterized by powder XRD. About 10 mol% of ceria could be dissolved in the lattice of ThSiO₄. A striking observation was the stabilization of tetragonal modification of ThSiO₄, which is metastable, by ceria substitution. There was no solubility of ceria in zircon (ZrSiO₄) lattice. The average linear thermal expansion coefficient (293–1123 K) of ZrSiO₄, ThSiO₄ and Th_0.9Ce_0.1SiO₄ are 4.65 × 10⁻⁶, 4.97 × 10⁻⁶ and 5.14 × 10⁻⁶ K⁻¹, respectively.