LPNN-based approach for LASSO problem via a sequence of regularized minimizations

In compressive sampling theory, the least absolute shrinkage and selection operator (LASSO) is a representative problem. Nevertheless, the non-differentiable constraint impedes the use of Lagrange programming neural networks (LPNNs). We present in this article the -LPNN model, a novel algorithm that tackles the LASSO minimization together with the underlying theory support. First, we design a sequence of smooth constrained optimization problems, by introducing a convenient differentiable approximation to the non-differentiable -norm constraint. Next, we prove that the optimal solutions of the regularized intermediate problems converge to the optimal sparse signal for the LASSO. Then, for every regularized problem from the sequence, the -LPNN dynamic model is derived, and the asymptotic stability of its equilibrium state is established as well. Finally, numerical simulations are carried out to compare the performance of the proposed -LPNN algorithm with both the LASSO-LPNN model and a standard digital method.     

Experimental characterization and performance study of an ammonia–water–hydrogen refrigerator

We present in this paper an experimental study of a commercial diffusion-absorption refrigeration machine (DAR) operating on the Platen and Munters cycle. The temperatures at the inlet and outlet of every component of the machine, as well as the cabinet and ambient temperature are measured continuously. The tests are repeated for various electric power inputs to the refrigerator. The global heat transfer coefficient of the cabinet (UA)_cab is determined using both theoretical and experimental methods. This coefficient is found equal to 0.2 W/°C. The global heat transfer coefficient of the evaporator (UA)_evap is deduced using dynamic and steady state methods. This global heat transfer coefficient (UA)_evap is found equal to 0.3 W/°C. Finally the cooling capacity of the unit and the coefficient of performance are evaluated. The heating power supply to the generator necessary to ensure the desired state of this machine is found to be in the range of 35 W–45 W.     

Modified smectitic Tunisian clays used in the formulation of high performance lubricating greases

Two smectitic Tunisian clays were organically modified by exchange reaction with dioctadecyl dimethylammonium chloride (DODMA) and hexadecyl benzyl dimethylammonium chloride (HBDMA). The resulting organoclays were used in the formulation of lubricating greases. The results of the performance tests applied to the obtained greases showed that their physicochemical and mechanical characteristics are in conformity with the extreme-pressure (EP) greases specifications. The obtained greases present good EP properties with only 0.5% EP additives. The smectitic clays used could constitute a raw material in the formulation of high performance lubricating greases.     

A comparison analysis of prefiltering methods in space-time trellis-based reduced-state turbo-detection

In this paper, we investigate two different prefiltering strategies, i.e., multiple input multiple output (MIMO) channel shortening and MIMO whitened matched filtering, to decode iteratively a space time bit interleaved coded modulation transmitted over MIMO block fading intersymbol interference channels using a trellis-based post-detector. Indeed, resorting to such front-ends allows significant complexity gain when compared to the optimal maximum a posteriori-based turbo-receiver. While those two approaches have been exhaustively visited in the past, their fair comparison in terms of trade-off between complexity and performance has – to our knowledge – never been tackled before. This paper aims at giving some valuable insights about the respective benefits of these two prefiltering methods with respect to the channel characteristics and overall receiver complexity.     

Optical properties of GaAs0.9-xNxSb0.1 alloy films studied by spectroscopic ellipsometry

Spectroscopic ellipsometry from 0.73 to 4.75 eV was used to study the optical properties of epitaxial GaAs_0.9-xN_xSb_0.1 layers with x = 0.00, 0.65, 1.06, 1.45 and 1.90%. The ellipsometric experimental spectra were fitted using a multilayer model employing the model dielectric function to describe the GaAs_0.9-xN_xSb_0.1 optical response. We have identified the Γ-point E₀, E₊, and E^# transitions of GaAs_0.9-xN_xSb_0.1 and have determined the effect of nitrogen on the respective transition energies. We have demonstrated that a lower N content can provide an equal E₊-E₀ energy splitting for GaAs_0.9-xN_xSb_0.1 with respect to GaAs_1-xN_x.     

Isolation of a yeast,Trichosporon cutaneum,able to use low molecular weight phenolic compounds: application to olive mill waste water treatment

This study deals with the degradation of phenolic compounds in olive oil mill waste; a highly polluting material in olive oil‐producing countries because of its abundance and the toxicity of its phenolic compound content. This investigation confirms the ability of an isolated yeast, identified as Trichosporon cutaneum, to degrade phenolic compounds extracted from olive mill waste water (OMW). The yeast was adapted to the OMW by an enrichment culture. The results of this biotransformation were a decrease in the phenolic content and hence a reduction in the phytotoxic effects of the effluent after the yeast treatment. The kinetic growth of the isolated yeast on phenol over a range of concentrations (0.3–3.0 g dm^?3) was studied. The ability of the strain to assimilate simple monomeric phenols and alkyl phenols, at a concentration of 1 g dm^?3, in a synthetic liquid medium used as the sole carbon source was investigated in a batch culture. The aromatic ring cleavage pathway occurred in the yeast through catechol oxidation. Using various concentrations of ethyl acetate extract from OMW as the sole carbon source, the yeast exhibited growth on the substrate up to 7 g dm^?3 equivalent of phenols. A significant reduction of COD after the treatment of the OMW extract by the yeast isolate was noticed. The removal of phenol and COD exceeded 80% of the original loading after 8 days of treatment, for extracts containing initial COD in the range 19 to 72 g dm^?3. Copyright © 2004 Society of Chemical Industry     

Two La0.5Ag0.1Ca0.4MnO3 Manganite Nanoparticles Synthesized via Sol–Gel and Solid-State Methods–Structural,Magnetic, and Magnetocaloric Properties

Journal of Superconductivity and Novel Magnetism - La0.5Ag0.1Ca0.4MnO3 manganite nanoparticles are synthesized via two different ways, namely, a solid-state reaction (S1) and the sol–gel...     

Elaboration and characterization of hybrid polymer electrolytes Nafion–TiO2 for PEMFCs

Nafion hybrid membrane containing titanium dioxide (TiO₂) nano-particles were elaborated by in situ sol–gel technique. Micro wave treatment was used during hybrid elaboration to 2 h. The X-ray diffraction (XRD) shows the crystallization of TiO₂ in anatase and rutile phases. The Infrared spectroscopy IR spectrum of hybrid membranes Nafion–TiO₂ confirm the presence of bonds Ti–O and Ti–OH which proves that the hybridization of the membrane by sol–gel method is successful. Atomic Force Microscopy AFM confirmed the good dispersion of TiO₂ nano-particles. The electrochemical impedance spectroscopy (EIS) and solvent uptake measurements suggest the enhancement of the membrane performance.     

Deflection Prediction of Steel and FRP-Reinforced Concrete-Filled FRP Tube Beams

This paper presents the results of experimental and theoretical investigations that study the flexural behavior of reinforced concrete-filled fiber-reinforced polymer (FRP) tubes (RCFFTs) beams. The experimental program consists of 10 circular beams [6 RCFFT and 4 control reinforced concrete (RC) beams] with a total length of 2,000?mm, tested under four-point bending load. The experimental results were used to review and verify the applicability of various North American code provisions and some available equations in the literature to predict deflection of RCFFT beams. The measured deflections and the experimental values of the effective moment of inertia were analyzed and compared with those predicted using available models. The results of the analysis indicated that the behavior of steel and FRP-RCFFT beams under the flexural load was significantly different than that of steel and FRP-RC members. This is attributed to the confining effect of the FRP tubes and their axial contribution. This confining behavior in turn enhanced the overall flexural behavior and improved the tension stiffening of RCFFT beams. For that, the predicted tension stiffening of steel and FRP-RCFFT beams using the conventional equations (steel or FRP-RC member) underestimates the flexural response; therefore, the predicted deflections are overestimated. Based on the analysis of the test results, the Branson’s equation for the effective moment of inertia of RC structures is modified, and new equations are developed to accurately predict the deflection of concrete-filled FRP tube (CFFT) beams reinforced with steel or FRP bars.