Optimization of artificial neural networks used for retention modelling in ion chromatography

The aim of this work is the development of an artificial neural network model, which can be generalized and used in a variety of applications for retention modelling in ion chromatography. Influences of eluent flow-rate and concentration of eluent anion (OH-) on separation of seven inorganic anions (fluoride, chloride, nitrite, sulfate, bromide, nitrate, and phosphate) were investigated. Parallel prediction of retention times of seven inorganic anions by using one artificial neural network was applied. MATLAB Neural Networks ToolBox was not adequate for application to retention modelling in this particular case. Therefore the authors adopted it for retention modelling by programming in MATLAB metalanguage. The following routines were written; the division of experimental data set on training and test set; selection of data for training and test set; Dixon's outlier test; retraining procedure routine; calculations of relative error. A three-layer feed forward neural network trained with a Levenberg-Marquardt batch error back propagation algorithm has been used to model ion chromatographic retention mechanisms. The advantage of applied batch training methodology is the significant increase in speed of calculation of algorithms in comparison with delta rule training methodology. The technique of experimental data selection for training set was used allowing improvement of artificial neural network prediction power. Experimental design space was divided into 8-32 subspaces depending on number of experimental data points used for training set. The number of hidden layer nodes, the number of iteration steps and the number of experimental data points used for training set were optimized. This study presents the very fast (300 iteration steps) and very accurate (relative error of 0.88%) retention model, obtained by using a small amount of experimental data (16 experimental data points in training set). This indicates that the method of choice for retention modelling in ion chromatography is the artificial neural network.     

Exact closed-form solutions for the natural frequencies and stability of elastically connected multiple beam system using Timoshenko and high-order shear deformation theory

A general procedure for the determination of the natural frequencies and buckling load for a set of beam system under compressive axial loading is investigated using Timoshenko and high-order shear deformation theory. It is assumed that the set beams of the system are simply supported and continuously joined by a Winkler elastic layer. The model of beam includes the effects of axial loading, shear deformation and rotary inertia. In the special case of identical beams, explicit expressions for the natural frequencies and the critical buckling load are determined using a trigonometric method. The influences of the compressive axial loading and the number of beams in the system on the natural frequencies and the critical buckling load are discussed. These results are of considerable practical interest and have wide application in engineering practice of frameworks.     

Spectrophotometric Study of Catechol Oxidation by Aerial O2 in Alkaline Aqueous Solutions Containing Mg(II) Ions.

Electronic absorption spectroscopy was employed to study the aerial oxidation of catechol (1,2-benzenediol) in alkaline aqueous solution containing an excess of Mg(II) ions. Graphical analysis by the matrix method of UV spectra recorded at regular time intervals gave a good fit for two absorbing species in solution. Based on this result and our earlier ESR spectroscopic investigations we concluded that two main absorbing species in this system are Mg(II)-spin stabilized o-benzosemiquinone anion radical and C-C dimer formed by the nucleophilic attack of catecholate anion on o-benzoquinone. Although the formation of 1,2,4-benzenetriol during the catechol oxidation has been detected in some ESR studies its presence was not indicated by this analysis probably because of the low and/or stable steady state concentration throughout the experiment.     

A new microcontroller-based RADFET dosimeter reader

A new reader for radiation dose measurements using RADFET (pMOSFET) dosemeters has been developed. The threshold voltage (V_T) of the pMOSFETs is measured using a “one-point” method that determines V_T as the gate voltage for a given drain current. Using V_T, the absorbed dose, which is directly proportional to the threshold voltage shift, is calculated. The reader is based on a low cost 8-bit PIC 18F4520 microcontroller (MCU), and works independently of a personal computer, uses a touch screen and stores the data in microcontroller memory. Good agreement in threshold voltage values, obtained using a high-quality source-measure unit and the reader, was obtained. In addition, the reader can be used for threshold voltage measurement with other types of MOSFETs, especially in long duration experiments, as well as for the real-time measurements in radiotherapy, either as an autonomous system or integrated in a larger monitoring configuration.     

微穿孔板结构在管道声源特性测量中的应用分析

将微穿孔板吸声结构应用到外加旁支管路的管道声源特性测量方法中,分析和仿真了单层及双层微穿孔板及其后腔结构采用不同参数(板厚、微孔直径、穿孔率、后腔深度等)时改变管道负载声阻抗的作用和影响.研究结果表明:改变板厚、微孔直径、穿孔率、管道横截面积比等参数,可以有效地改变负载声阻;改变后腔深度及微穿孔板与主体管道之间距离,可...     

Magneto-intersubband zener tunneling in a wide GaAs quantum well at high filling factors

The dependence of the differential resistance r _xx on the dc current density J _dc in a wide GaAs quantum well with two occupied size quantization subbands has been investigated at the temperature T = 4.2 K in the magnetic fields B < 1 T. A peak, whose position is given by the relation 2R _c eE _H = ħω_c/2, where R _c is the cyclotron radius, E _H is the Hall electric field, and ω_c is the cyclotron frequency, has been observed in the r _xx (J _dc) curves at high filling factors. The experimental results are attributed to Zener tunneling of electrons between the Landau levels of different subbands.     

Dependences of the Transport Scattering Time and Quantum Lifetime on the Two-Dimensional Electron Gas Density in Modulation-Doped Single GaAs Quantum Wells with AlAs/GaAs Short-Period Superlattice Barriers

JETP Letters - The dependences of the transport scattering time τt, quantum lifetime τq, and their ratio τt/τq on the density ne of the electron gas in modulation-doped single...     

Signature of the conformational preferences of small peptides: a theoretical investigation

An extensive computational study of the conformational preferences of N-acetylphenylalaninylamide (NAPA) is reported, including conformational and anharmonic frequency analyses, as well as calculations of excitation energies of the four NAPA conformers lowest in energy. Particular attention is paid to the influence of hydrogen-bonding interactions on the relative stability of the conformers, which was found to be very sensitive to both the level of quantum chemical computations and the anharmonic treatment of molecular vibrations. The assignments of the UV spectral peaks are well supported by the multireference CASSCF/MS-CASPT2 calculations. Upon consideration of the second-order M?ller-Plesset (MP2) and density functional theory (DFT) structures, overall energetics, and harmonic and anharmonic corrections, we found no conclusive theoretical evidence for the assumed conformational propensity of small model peptides toward extended beta-strand structures.     

Synthesis of novel ferrocene-containing 1,3-thiazinan-2-imines: One-pot reaction promoted by ultrasound irradiation

A simple one-pot synthesis of new ferrocene-containing 1,3-thiazinan-2-imines from 3-arylamino-1-ferrocenylpropan-1-ols and phenyl isothiocyanate has been developed. The key intermediate β-hydroxy thioureas were generated in situ using ultrasound irradiation and subsequent cyclization was achieved by the addition of acetic acid. The scope of the reaction towards various 3-arylamino-1-ferrocenylpropan-1-ols has been explored and the corresponding 3-aryl-6-ferrocenyl-N-phenyl-1,3-thiazinan-2-imines were obtained in moderate to high yields (52–90%).     

Effect of parameters of square-wave pulsating current on copper electrodeposition in the hydrogen co-deposition range

Copper electrodeposition processes in the hydrogen co-deposition range by the pulsating current (PC) regime were examined by the determination of the average current efficiency of hydrogen evolution and by the scanning electron microscopic (SEM) analysis of the morphology of the formed deposits. The two sets of the square-wave PC of the same pause to pulse ratios, but with different duration of deposition pulses and pauses were analyzed. The one set of square-wave PC was with the constant pause duration and different deposition pulses. In the other set, the deposition pulse was constant while the pause duration was varied. The obtained results were compared with those obtained by electrodeposition at the constant overpotential from solutions of different CuSO₄ and H₂SO₄ concentrations. It was found that the effect of the increasing deposition pulse was equivalent to the decreasing CuSO₄ concentration, while the effect of the decreasing pause duration was equivalent to the increasing H₂SO₄ concentration. It is shown that it is possible to achieve a substitution of more solutions of different CuSO₄ and H₂SO₄ concentartions by the use of the only one solution if the appropriate PC parameters were applied, what can be of high technological significance.