Particle-in-Cell algorithms for emerging computer architectures

We have designed Particle-in-Cell algorithms for emerging architectures. These algorithms share a common approach, using fine-grained tiles, but different implementations depending on the architecture. On the GPU, there were two different implementations, one with atomic operations and one with no data collisions, using CUDA C and Fortran. Speedups up to about 50 compared to a single core of the Intel i7 processor have been achieved. There was also an implementation for traditional multi-core processors using OpenMP which achieved high parallel efficiency. We believe that this approach should work for other emerging designs such as Intel Phi coprocessor from the Intel MIC architecture.     

ESTIMATION OF VAPOR PRESSURES OF ORGANIC NITROGEN AND SULFUR COMPOUNDS BY A GROUP-CONTRIBUTION METHOD

The group-contribution method for vapor pressures of hydrocarbons and organic compounds containing nitrogen or sulfur, based on the kinetic theory of fluids, is revised and extended to include new groups containing nitrogen or sulfur. Good representation is obtained for vapor pressure data in the region 1.30-270 kPa. The method may be used to estimate vapor pressures and enthalpies of vaporization for those organic fluids containing nitrogen or sulfur, where no experimental data are available.     

A hybrid evolutionary approach to the registration area planning problem

The last two decades have witnessed a phenomenal growth in the number of cellular wireless network users which in turn stressed the need to utilize the limited network bandwidth in an efficient manner. The network bandwidth is consumed not only by user traffic, but also by control traffic needed for ensuring the mobility of users. As we don’t have any control over the volume of user traffic, all attempts to efficiently use bandwidth are based on frequency reuse and minimizing the control traffic. The registration area planning (RAP) problem seeks a partition of the cells of the network into contiguous areas called registration areas so that the bandwidth consumed by control signals is minimized. RAP problem in an \(\mathcal {NP}\) -Hard problem. In this paper, we present a steady-state grouping genetic algorithm with local search to solve this problem. We have compared our approach with the state-of-the-art approaches reported in the literature. Computational results show the effectiveness of our approach.     

Response of wheat (Triticum aestivum) and pearl-millet (Pennistum americanum) to zinc fertilizers on two soils

Responses of ZnSO₄, Zn₃ (PO₄)₂ · 4H₂O, ZnO, ZnNH₄PO₃ · 2H₂O and zincated superphosphate were studied on wheat (Triticum aestivum) and pearl-millet (Pennistem americanum) in Balsamand sand (Ustipsament) and Ladwa fine loam (Typic Camborthids). In Balsamand sand all Zn fertilizers increased wheat and pearl-milllet grain yield significantly over control but ZnNH₄ PO₄ · 2H₂O and zincated super were the best sources. Straw yield was also highest due to zincated super and ZnNH₄PO₄ · 2H₂O applications. Zinc sulphate was significantly inferior to zincated super and ZnNH₄PO₄ · 2H₂O. In Ladwa fine loam, all fertilizers gave significantly higher grain yield of wheat and pearl-millet than control except ZnO in pearl-millet. Highest yield in this soil was obtained by ZnSO₄ · 7H₂O followed by ZnNH₄ PO₄ · 2H₂O and then zincated super.In Balsamand sand, the Zn fertilizers significantly increased the Zn concentration and Zn uptake of wheat grain. In pearl-millet, only Zn uptake was increased significantly with Zn fertilizers. Zincated super gave highest Zn uptake.The concentration of P in wheat grain was increased by Zn fertilizers in Balsamand sand, and also in pearl-millet where zincated super and ZnNH₄PO₄ were most effective in increasing the P content.When Zn fertilizers were applied to preceding pearl-millet the effect on succeeding wheat crop were in the same order as direct application of Zn fertilizers. Zincated super was the best and ZnO worst with respect to wheat grain and straw yield and uptake of Zn and P.     

Hydration Study of the System Ca3Al2O6–CaSO4· 2H2O–Ca(OH)2–H2O with and without Citric Acid

Hydration occurring in the system Ca₃Al₂O₆–CaSO₄· 2H₂O–Ca(OH)₂–H₂O has been studied at different temperatures and it was found that the reactions are diffusion controlled. The kinetic data obeyed Jander's equation and the rate of reaction increased with increasing temperature. X-ray diffraction studies and calorimetric measurements show that when gypsum is consumed, ettringite is converted into monosulfate. The rate of this conversion also increased with the increasing temperature and decreased in the presence of citric acid. Spectroscopic studies showed that there was some interaction between citric acid and the cement and that the product of hydration is of colloidal nature. Zeta potential measurements show that retardation of Ca₃Al₂O₆ hydration in the presence of gypsum and Ca(OH)₂ is not due to SO²⁻₄ adsorption. Electrical conductivity and thermoelectric potential measurements of solid Ca₃Al₂O₆ show that Ca₃Al₂O₆ is an n -type semiconductor and contains defects. The retardation of Ca₃Al₂O₆ may be due to poisoning of reaction sites by gypsum and Ca(OH)_2.     

Diffusion coefficient in polymer solutions

Molecular diffusivity of a solute in a solvent may be determined by measuring the extent of dispersion of solute in solvent flowing in a straight circular tube under the conditions of laminar flow. This simple and rapid method for determination of molecular diffusivity in aquous polymer solutions is discussed. Experimental results show a substantial reduction in the solute diffusivity with increase in polymer concentration.     

Predicting external sulphur requirement of alfalfa from sulphate sorption isotherm of Tanzanian soils

For Tanzanian soils dominant in hydrous oxides of iron and amorphous ferri-alumino silicate, a 48-hour (hr) mixing period with the sulphate (SO₄) solution was adequate for a near-equilibrium condition. Although differing in their SO₄ sorption capacity, all the soils sorbed SO₄ at or beyond 1µg ml^–1 sulphur (S) concentration in the supernatant. Hydroxyl (OH) ions were displaced during SO₄ sorption as indicated by a significant positive correlation between the amount of sorbed SO₄ and the difference in pH values determined in 0.1N K₂ SO₄ and 0.1N KCl, i.e. the dpH values.In a greenhouse experiment, alfalfa was grown on eight soils at six adjusted S concentrations. Sulphur deficiency symptoms appeared in the control pots of those soils which were low in native sorbed SO₄, SO₄ sorption capacity and initial soil solution S concentration. Sulphur fertilization increased dry matter (DM) yield as well as response to applied S. The external S concentration, i.e. adjusted S concentration required for 95% of the maximum DM yield, ranged from 0.8 to 8.2µg S ml^–1 with values less than 2.0 on most of the soils. The external S concentration decreased hyperbolically as the SO₄ sorption capacity of the soils increased. The total amount of fertilizer S required to obtain the external S concentration in solution, and at the same time satisfy the SO₄ sorption capacity of the soil at the external S concentration (determined from the sorption isotherm) was defined as the external S requirement for the specified yield level of alfalfa. The external S requirement for 95% of the maximum yield of alfalfa varied from soil to soil due to differences in their capacity and intensity for S nutrition.Part of a thesis by the senior author for the MSc (Agric) degree of the University of Dar es Salaam     

An efficient approach for face recognition based on common eigenvalues

In this paper, a simple technique is proposed for face recognition among many human faces. It is based on the polynomial coefficients, covariance matrix and algorithm on common eigenvalues. The main advantage of the proposed approach is that the identification of similarity between human faces is carried out without computing actual eigenvalues and eigenvectors. A symmetric matrix is calculated using the polynomial coefficients-based companion matrices of two compared images. The nullity of a calculated symmetric matrix is used as similarity measure for face recognition. The value of nullity is very small for dissimilar images and distinctly large for similar face images. The feasibility of the propose approach is demonstrated on three face databases, i.e., the ORL database, the Yale database B and the FERET database. Experimental results have shown the effectiveness of the proposed approach for feature extraction and classification of the face images having large variation in pose and illumination.     

The C-loss function for pattern classification

This paper presents a new loss function for neural network classification, inspired by the recently proposed similarity measure called Correntropy. We show that this function essentially behaves like the conventional square loss for samples that are well within the decision boundary and have small errors, and L₀ or counting norm for samples that are outliers or are difficult to classify. Depending on the value of the kernel size parameter, the proposed loss function moves smoothly from convex to non-convex and becomes a close approximation to the misclassification loss (ideal 0–1 loss). We show that the discriminant function obtained by optimizing the proposed loss function in the neighborhood of the ideal 0–1 loss function to train a neural network is immune to overfitting, more robust to outliers, and has consistent and better generalization performance as compared to other commonly used loss functions, even after prolonged training. The results also show that it is a close competitor to the SVM. Since the proposed method is compatible with simple gradient based online learning, it is a practical way of improving the performance of neural network classifiers.     

Surface tension and contact angle of herbicide solutions affected by surfactants

Contact angle and surface tension were measured for distilled and hard water solutions of adjuvants, Ortho X-77, Span-20, Sterox-NJ. Surfactant-WK, Triton B-1956, Triton X-114, Tween-20, and Sun Oil 11E. The same parameters were measured for suspensions of atrazine [2-chloro-4-(ethylamino)-6-(isopropylamino)-s-triazine] and ametryne [2-(ethylamino)-4-(isopropylamino)-6-(methylthio)-s-triazine] with and without each adjuvant. All adjuvants reduced surface tension and contact angle of distilled water; Surfactant-WK was most effective and Tween-20 was least effective. Increasing concentration of surfactants from 0 to 0.1% (v/v) gave progressive reduction in surface tension and contact angle while higher concentrations, 0.1 to 2.0% (v/v), had no further effect. Surfactant-WK at 0.1% (v/v) in distilled water reduced the surfact tension from 72.8 dynes/cm to 27 dynes/cm and contact angle from 110° to 41°. An additional increase in Surfactant-WK concentration from 0.1% (v/v) to 2% (v/v) did not further reduce surface tension and contact angle. Sun Oil 11E was identical in behavior except that it was less effective than the surfactants. Water hardness up to 1,000 ppm as Ca ions did not affect surface tension and contact angle in surfactant solutions. An aqueous solution of atrazine had a higher surface tension and contact angle than ametryne in the absence of surfactants. However, these differences were not observed when surfactants were added to either herbicide.