Describing function of two masses with backlash

This paper analyzes the dynamical properties of systems with backlash and impact phenomena based on the describing function method. It is shown that this type of nonlinearity can be analyzed in the perspective of the fractional calculus theory. The fractional dynamics is compared with that of standard models.

Modeling reaction kinetics and mass transfer in ozonation in water solutions

This dissertation is based on four articles dealing with modeling of ozonation. The literature part of this considers some models for hydrodynamics in bubble column simulation. A literature review of methods for obtaining mass transfer coefficients is presented. The methods presented to obtain mass transfer are general models and can be applied to any gas-liquid system. Ozonation reaction models and methods for obtaining stoichiometric coefficients and reaction rate coefficients for ozonation reactions are discussed in the final section of the literature part. In the first article, ozone gas-liquid mass transfer into water in a bubble column was investigated for different pH values. A more general method for estimation of mass transfer and Henry’s coefficient was developed from the Beltrán method. The ozone volumetric mass transfer coefficient and the Henry’s coefficient were determined simultaneously by parameter estimation using a nonlinear optimization method. A minor dependence of the Henry’s law constant on pH was detected at the pH range 4 - 9. In the second article, a new method using the axial dispersion model for estimation of ozone self-decomposition kinetics in a semi-batch bubble column reactor was developed. The reaction rate coefficients for literature equations of ozone decomposition and the gas phase dispersion coefficient were estimated and compared with the literature data. The reaction order in the pH range 7-10 with respect to ozone 1.12 and 0.51 the hydroxyl ion were obtained, which is in good agreement with literature. The model parameters were determined by parameter estimation using a nonlinear optimization method. Sensitivity analysis was conducted using object function method to obtain information about the reliability and identifiability of the estimated parameters. In the third article, the reaction rate coefficients and the stoichiometric coefficients in the reaction of ozone with the model component p-nitrophenol were estimated at low pH of water using nonlinear optimization. A novel method for estimation of multireaction model parameters in ozonation was developed. In this method the concentration of unknown intermediate compounds is presented as a residual COD (chemical oxygen demand) calculated from the measured COD and the theoretical COD for the known species. The decomposition rate of p-nitrophenol on the pathway producing hydroquinone was found to be about two times faster than the p-nitrophenol decomposition rate on the pathway producing 4- nitrocatechol. In the fourth article, the reaction kinetics of p-nitrophenol ozonation was studied in a bubble column at pH 2. Using the new reaction kinetic model presented in the previous article, the reaction kinetic parameters, rate coefficients, and stoichiometric coefficients as well as the mass transfer coefficient were estimated with nonlinear estimation. The decomposition rate of pnitrophenol was found to be equal both on the pathway producing hydroquinone and on the path way producing 4-nitrocathecol. Comparison of the rate coefficients with the case at initial pH 5 indicates that the p-nitrophenol degradation producing 4- nitrocathecol is more selective towards molecular ozone than the reaction producing hydroquinone. The identifiability and reliability of the estimated parameters were analyzed with the Marcov chain Monte Carlo (MCMC) method. @All rights reserved. No part of the publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the author.

Energy partitioning for "fuzzy" atoms

The total energy of molecule in terms of 'fuzzy atoms' presented as sum of one- and two-atomic energy components is described. The divisions of three-dimensional physical space into atomic regions exhibit continuous transition from one to another. The energy components are on chemical energy scale according to proper definitions. The Becke's integration scheme and weight function determines realization of method which permits effective numerical integrations

Equidistributing Meshes with Constraints

Adaptive methods which “equidistribute” a given positive weight function are now used fairly widely for selecting discrete meshes. The disadvantage of such schemes is that the resulting mesh may not be smoothly varying. In this paper a technique is developed for equidistributing a function subject to constraints on the ratios of adjacent steps in the mesh. Given a weight function $f \geqq 0$ on an interval $[a,b]$ and constants $c$ and $K$, the method produces a mesh with points $x_0 = a,x_{j + 1} = x_j + h_j ,j = 0,1, \cdots ,n - 1$ and $x_n = b$ such that\[ \int_{xj}^{x_{j + 1} } {f \leqq c\quad {\text{and}}\quad \frac{1} {K}} \leqq \frac{{h_{j + 1} }} {{h_j }} \leqq K\quad {\text{for}}\, j = 0,1, \cdots ,n - 1 . \] A theoretical analysis of the procedure is presented, and numerical algorithms for implementing the method are given. Examples show that the procedure is effective in practice. Other types of constraints on equidistributing meshes are also discussed. The principal application of the procedure is to the solution of boundary value problems, where the weight function is generally some error indicator, and accuracy and convergence properties may depend on the smoothness of the mesh. Other practical applications include the regrading of statistical data.

THE PARTITION-FUNCTION FOR AN ANYON-LIKE OSCILLATOR

We compute the partition function of an anyon-like harmonic oscillator. The well known results for both the bosonic and fermionic oscillators are then re-obtained as particular cases of our function. The technique we employ is a non-relativistic version of the Green function method used in the computation of one-loop effective actions of quantum field theory.

Three-dimensional finite element analysis of MHD duct flow by the penalty function formulation

A numerical scheme based on the Finite Element Method (FEM) is presented to calculate the full solution of a three-dimensional steady magnetohydrodynamic (MHD) flow with moderately high Hartmann numbers and interaction parameters. An incompressible, viscous and electrically conducting liquid-metal is considered. Assuming a low magnetic Reynolds number, the solution method solves the coupled Navier-Stokes and Maxwell's equations through the use of a penalty function method. Results are presented for Hartmann numbers in the range 10(2)-10(3).

Particle competition and cooperation to prevent error propagation from mislabeled data in semi-supervised learning

Semi-supervised learning is applied to classification problems where only a small portion of the data items is labeled. In these cases, the reliability of the labels is a crucial factor, because mislabeled items may propagate wrong labels to a large portion or even the entire data set. This paper aims to address this problem by presenting a graph-based (network-based) semi-supervised learning method, specifically designed to handle data sets with mislabeled samples. The method uses teams of walking particles, with competitive and cooperative behavior, for label propagation in the network constructed from the input data set. The proposed model is nature-inspired and it incorporates some features to make it robust to a considerable amount of mislabeled data items. Computer simulations show the performance of the method in the presence of different percentage of mislabeled data, in networks of different sizes and average node degree. Importantly, these simulations reveals the existence of the critical points of the mislabeled subset size, below which the network is free of wrong label contamination, but above which the mislabeled samples start to propagate their labels to the rest of the network. Moreover, numerical comparisons have been made among the proposed method and other representative graph-based semi-supervised learning methods using both artificial and real-world data sets. Interestingly, the proposed method has increasing better performance than the others as the percentage of mislabeled samples is getting larger. © 2012 IEEE.

Obtenção das distribuições de carga/massa e momentos no 'INTPOT. 4 HE' com o método de coordenadas geradoras

Pós-graduação em Física - IFT

Migração (2,5-D) com amplitudes verdadeiras em meios com gradiente constante de velocidade

A migração com amplitudes verdadeiras de dados de reflexão sísmica, em profundidade ou em tempo, possibilita que seja obtida uma medida dos coeficientes de reflexão dos chamados eventos de reflexão primária. Estes eventos são constituídos, por exemplo, pelas reflexões de ondas longitudinais P-P em refletores de curvaturas arbitrárias e suaves. Um dos métodos mais conhecido é o chamado migração de Kirchhoff, através do qual a imagem sísmica é produzida pela integração do campo de ondas sísmicas, utilizando-se superfícies de difrações, denominadas de Superfícies de Huygens. A fim de se obter uma estimativa dos coeficientes de reflexão durante a migração, isto é a correção do efeito do espalhamento geométrico, utiliza-se uma função peso no operador integral de migração. A obtenção desta função peso é feita pela solução assintótica da integral em pontos estacionários. Tanto no cálculo dos tempos de trânsito como na determinação da função peso, necessita-se do traçamento de raios, o que torna a migração em situações de forte heterogeneidade da propriedade física um processo com alto custo computacional. Neste trabalho é apresentado um algoritmo de migração em profundidade com amplitudes verdadeiras, para o caso em que se tem uma fonte sísmica pontual, sendo o modelo de velocidades em subsuperfície representado por uma função que varia em duas dimensões, e constante na terceira dimensão. Esta situação, conhecida como modelo dois-e-meio dimensional (2,5-D), possui características típicas de muitas situações de interesse na exploração do petróleo, como é o caso da aquisição de dados sísmicos 2-D com receptores ao longo de uma linha sísmica e fonte sísmica 3-D. Em particular, é dada ênfase ao caso em que a velocidade de propagação da onda sísmica varia linearmente com a profundidade. Outro tópico de grande importância abordado nesse trabalho diz respeito ao método de inversão sísmica denominado empilhamento duplo de difrações. Através do quociente de dois empilhamentos com pesos apropriados, pode-se determinar propriedades físicas e parâmetros geométricos relacionados com a trajetória do raio refletido, os quais podem ser utilizados a posteriori no processamento dos dados sísmicos, visando por exemplo, a análise de amplitudes.

Diagnostic accuracy of semi-quantitative and quantitative culture techniques for the diagnosis of catheter-related infections in newborns and molecular typing of isolated microorganisms

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Combined active and semi-supervised learning using particle walking temporal dynamicscombined active and semi-supervised learning using particle walking temporal dynamics

Both Semi-Supervised Leaning and Active Learning are techniques used when unlabeled data is abundant, but the process of labeling them is expensive and/or time consuming. In this paper, those two machine learning techniques are combined into a single nature-inspired method. It features particles walking on a network built from the data set, using a unique random-greedy rule to select neighbors to visit. The particles, which have both competitive and cooperative behavior, are created on the network as the result of label queries. They may be created as the algorithm executes and only nodes affected by the new particles have to be updated. Therefore, it saves execution time compared to traditional active learning frameworks, in which the learning algorithm has to be executed several times. The data items to be queried are select based on information extracted from the nodes and particles temporal dynamics. Two different rules for queries are explored in this paper, one of them is based on querying by uncertainty approaches and the other is based on data and labeled nodes distribution. Each of them may perform better than the other according to some data sets peculiarities. Experimental results on some real-world data sets are provided, and the proposed method outperforms the semi-supervised learning method, from which it is derived, in all of them.

Semi-supervised learning guided by the modularity measure in complex networks

Semi-supervised learning techniques have gained increasing attention in the machine learning community, as a result of two main factors: (1) the available data is exponentially increasing; (2) the task of data labeling is cumbersome and expensive, involving human experts in the process. In this paper, we propose a network-based semi-supervised learning method inspired by the modularity greedy algorithm, which was originally applied for unsupervised learning. Changes have been made in the process of modularity maximization in a way to adapt the model to propagate labels throughout the network. Furthermore, a network reduction technique is introduced, as well as an extensive analysis of its impact on the network. Computer simulations are performed for artificial and real-world databases, providing a numerical quantitative basis for the performance of the proposed method.

Exact correlation functions in particle-reaction models with immobile particles

Exact results on particle densities as well as correlators in two models of immobile particles, containing either a single species or else two distinct species, are derived. The models evolve following a descent dynamics through pair annihilation where each particle interacts once at most throughout its entire history. The resulting large number of stationary states leads to a non-vanishing configurational entropy. Our results are established for arbitrary initial conditions and are derived via a generating function method. The single-species model is the dual of the 1D zero-temperature kinetic Ising model with Kimball-Deker-Haake dynamics. In this way, both in finite and semi-infinite chains and also the Bethe lattice can be analysed. The relationship with the random sequential adsorption of dimers and weakly tapped granular materials is discussed.

Body contouring surgery following bariatric surgery and dietetically induced massive weight reduction: a risk analysis

BACKGROUND: This study analyzed the impact of weight reduction method, preoperative, and intraoperative variables on the outcome of reconstructive body contouring surgery following massive weight reduction. METHODS: All patients presenting with a maximal BMI >/=35 kg/m(2) before weight reduction who underwent body contouring surgery of the trunk following massive weight loss (excess body mass index loss (EBMIL) >/= 30%) between January 2002 and June 2007 were retrospectively analyzed. Incomplete records or follow-up led to exclusion. Statistical analysis focused on weight reduction method and pre-, intra-, and postoperative risk factors. The outcome was compared to current literature results. RESULTS: A total of 104 patients were included (87 female and 17 male; mean age 47.9 years). Massive weight reduction was achieved through bariatric surgery in 62 patients (59.6%) and dietetically in 42 patients (40.4%). Dietetically achieved excess body mass index loss (EBMIL) was 94.20% and in this cohort higher than surgically induced reduction EBMIL 80.80% (p < 0.01). Bariatric surgery did not present increased risks for complications for the secondary body contouring procedures. The observed complications (26.9%) were analyzed for risk factors. Total tissue resection weight was a significant risk factor (p < 0.05). Preoperative BMI had an impact on infections (p < 0.05). No impact on the postoperative outcome was detected in EBMIL, maximal BMI, smoking, hemoglobin, blood loss, body contouring technique or operation time. Corrective procedures were performed in 11 patients (10.6%). The results were compared to recent data. CONCLUSION: Bariatric surgery does not increase risks for complications in subsequent body contouring procedures when compared to massive dietetic weight reduction.

Benthic and planktonic foraminifera of sediment core GIK17957-2

A detailed age model for core 17957-2 of the southern South China Sea was developed based on delta18O, coarse fraction, magnetostratigraphy, and biostratigraphy for the last 1500 kyr. The delta18O record has clear ~100-kyr cycles after the Mid-Pleistocene Revolution (MPR) at the entrance of marine isotopic stage (MIS) 22. Planktonic foraminifera responded to the MPR immediately, showing the increased sea surface temperature (SST) and dissolution after the MPR. Benthic foraminifera did not respond to it until the Brunhes/Matuyama boundary. Since the MPR, the depth of thermocline gradually became shallower until MISs 6-5. This major change within MISs 6-5 was also reflected in the decreased SSTs and increased productivity and Deep Water Mass. Thus two major Pleistocene paleoceanographic changes were found: One was around the MPR; the other occurred within MISs 6-5, which speculatively might be ascribed to the reorganization of surface and deep circulation, possibly induced by tectonic forces.