Hybrid Water Demand Forecasting Model Associating Artificial Neural Network with Fourier Series

This paper addressed the problem of water-demand forecasting for real-time operation of water supply systems. The present study was conducted to identify the best fit model using hourly consumption data from the water supply system of Araraquara, Sa approximate to o Paulo, Brazil. Artificial neural networks (ANNs) were used in view of their enhanced capability to match or even improve on the regression model forecasts. The ANNs used were the multilayer perceptron with the back-propagation algorithm (MLP-BP), the dynamic neural network (DAN2), and two hybrid ANNs. The hybrid models used the error produced by the Fourier series forecasting as input to the MLP-BP and DAN2, called ANN-H and DAN2-H, respectively. The tested inputs for the neural network were selected literature and correlation analysis. The results from the hybrid models were promising, DAN2 performing better than the tested MLP-BP models. DAN2-H, identified as the best model, produced a mean absolute error (MAE) of 3.3 L/s and 2.8 L/s for training and test set, respectively, for the prediction of the next hour, which represented about 12% of the average consumption. The best forecasting model for the next 24 hours was again DAN2-H, which outperformed other compared models, and produced a MAE of 3.1 L/s and 3.0 L/s for training and test set respectively, which represented about 12% of average consumption. DOI: 10.1061/(ASCE)WR.1943-5452.0000177. (C) 2012 American Society of Civil Engineers.

Fog- and cloud-induced aerosol modification observed by the Aerosol Robotic Network (AERONET)

Large fine mode-dominated aerosols (submicron radius) in size distributions retrieved from the Aerosol Robotic Network (AERONET) have been observed after fog or low-altitude cloud dissipation events. These column-integrated size distributions have been obtained at several sites in many regions of the world, typically after evaporation of low-altitude cloud such as stratocumulus or fog. Retrievals with cloud-processed aerosol are sometimes bimodal in the accumulation mode with the larger-size mode often similar to 0.4-0.5 mu m radius (volume distribution); the smaller mode, typically similar to 0.12 to similar to 0.20 mu m, may be interstitial aerosol that were not modified by incorporation in droplets and/or aerosol that are less hygroscopic in nature. Bimodal accumulation mode size distributions have often been observed from in situ measurements of aerosols that have interacted with clouds, and AERONET size distribution retrievals made after dissipation of cloud or fog are in good agreement with particle sizes measured by in situ techniques for cloud-processed aerosols. Aerosols of this type and large size range (in lower concentrations) may also be formed by cloud processing in partly cloudy conditions and may contribute to the "shoulder" of larger-size particles in the accumulation mode retrievals, especially in regions where sulfate and other soluble aerosol are a significant component of the total aerosol composition. Observed trends of increasing aerosol optical depth (AOD) as fine mode radius increased suggests higher AOD in the near-cloud environment and higher overall AOD than typically obtained from remote sensing owing to bias toward sampling at low cloud fraction.

Structure and Ionic Conductivity in the Mixed-Network Former Chalcogenide Glass System [Na2S](2/3)[(B2S3)(x)(P2S5)(1-x)](1/3)

Glasses in the system [Na2S](2/3)[(B2S3)(x)(P2S5)(1-x)](1/3) (0.0 <= x <= 1.0) were prepared by the melt quenching technique, and their properties were characterized by thermal analysis and impedance spectroscopy. Their atomic-level structures were comprehensively characterized by Raman spectroscopy and B-11, P-31, and Na-23 high resolution solid state magic-angle spinning (MAS) NMR techniques. P-31 MAS NMR peak assignments were made by the presence or absence of homonuclear indirect P-31-P-31 spin-spin interactions as detected using homonuclear J-resolved and refocused INADEQUATE techniques. The extent of B-S-P connectivity in the glassy network was quantified by P-31{B-11} and B-11{P-31} rotational echo double resonance spectroscopy. The results clearly illustrate that the network modifier alkali sulfide, Na2S, is not proportionally shared between the two network former components, B and P. Rather, the thiophosphate (P) component tends to attract a larger concentration of network modifier species than predicted by the bulk composition, and this results in the conversion of P2S74-, pyrothiophosphate, Na/P = 2:1, units into PS43-, orthothiophosphate, Na/P = 3:1, groups. Charge balance is maintained by increasing the net degree of polymerization of the thioborate (B) units through the formation of covalent bridging sulfur (BS) units, B S B. Detailed inspection of the B-11 MAS NMR spectra reveals that multiple thioborate units are formed, ranging from neutral BS3/2 groups all the way to the fully depolymerized orthothioborate (BS33-) species. On the basis of these results, a comprehensive and quantitative structural model is developed for these glasses, on the basis of which the compositional trends in the glass transition temperatures (T-g) and ionic conductivities can be rationalized. Up to x = 0.4, the dominant process can be described in a simplified way by the net reaction equation P-1 + B-1 reversible arrow P-0 + B-4, where the superscripts denote the number of BS atoms for the respective network former species. Above x = 0.4, all of the thiophosphate units are of the P-0 type and both pyro-(B-1) and orthothioborate (B-0) species make increasing contributions to the network structure with increasing x. In sharp contrast to the situation in sodium borophosphate glasses, four-coordinated thioborate species are generally less abundant and heteroatomic B-S-P linkages appear to not exist. On the basis of this structural information, compositional trends in the ionic conductivities are discussed in relation to the nature of the charge-compensating anionic species and the spatial distribution of the charge carriers.

LARGE-SCALE DISTRIBUTION OF ARRIVAL DIRECTIONS OF COSMIC RAYS DETECTED ABOVE 10(18) eV AT THE PIERRE AUGER OBSERVATORY

A thorough search for large-scale anisotropies in the distribution of arrival directions of cosmic rays detected above 10(18) eV at the Pierre Auger Observatory is presented. This search is performed as a function of both declination and right ascension in several energy ranges above 10(18) eV, and reported in terms of dipolar and quadrupolar coefficients. Within the systematic uncertainties, no significant deviation from isotropy is revealed. Assuming that any cosmic-ray anisotropy is dominated by dipole and quadrupole moments in this energy range, upper limits on their amplitudes are derived. These upper limits allow us to test the origin of cosmic rays above 10(18) eV from stationary Galactic sources densely distributed in the Galactic disk and predominantly emitting light particles in all directions.

Quantification of Short and Medium Range Order in Mixed Network Former Glasses of the System GeO2-NaPO3: A Combined NMR and X-ray Photoelectron Spectroscopy Study

Glasses in the system xGeO(2)-(1-x)NaPO3 (0 <= x <= 0.50) were prepared by conventional melting quenching and characterized by thermal analysis, Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), and P-31 nuclear magnetic resonance (MAS NMR) techniques. The deconvolution of the latter spectra was aided by homonuclear J-resolved and refocused INADEQUATE techniques. The combined analyses of P-31 MAS NMR and O-1s XPS lineshapes, taking charge and mass balance considerations into account, yield the detailed quantitative speciations of the phosphorus, germanium, and oxygen atoms and their respective connectivities. An internally consistent description is possible without invoking the formation of higher-coordinated germanium species in these glasses, in agreement with experimental evidence in the literature. The structure can be regarded, to a first approximation, as a network consisting of P-(2) and P-(3) tetrahedra linked via four-coordinate germanium. As implied by the appearance of P-(3) units, there is a moderate extent of network modifier sharing between phosphate and germanate network formers, as expressed by the formal melt reaction P-(2) + Ge-(4) -> P-(3) + Ge-(3). The equilibrium constant of this reaction is estimated as K = 0.52 +/- 0.11, indicating a preferential attraction of network modifier by the phosphorus component. These conclusions are qualitatively supported by Raman spectroscopy as well as P-31{Na-23} and P-31{Na-23} rotational echo double resonance (REDOR) NMR results. The combined interpretation of O-1s XPS and P-31 MAS NMR spectra shows further that there are clear deviations from a random connectivity scenario: heteroatomic P-O-Ge linkages are favored over homoatomic P-O-P and Ge-O-Ge linkages.

Groups of services delivered by Brazilian branchless banking and respective network integration models

Over the last decade, Brazil has pioneered an innovative model of branchless banking, known as correspondent banking, involving distribution partnership between banks, several kinds of retailers and a variety of other participants, which have allowed an unprecedented growth in bank outreach and became a reference worldwide. However, despite the extensive number of studies recently developed focusing on Brazilian branchless banking, there exists a clear research gap in the literature. It is still necessary to identify the different business configurations involving network integration through which the branchless banking channel can be structured, as well as the way they relate to the range of bank services delivered. Given this gap, our objective is to investigate the relationship between network integration models and services delivered through the branchless banking channel. Based on twenty interviews with managers involved with the correspondent banking business and data collected on almost 300 correspondent locations, our research is developed in two steps. First, we created a qualitative taxonomy through which we identified three classes of network integration models. Second, we performed a cluster analysis to explain the groups of financial services that fit each model. By contextualizing correspondents' network integration processes through the lens of transaction costs economics, our results suggest that the more suited to deliver social-oriented, "pro-poor'' services the channel is, the more it is controlled by banks. This research offers contributions to managers and policy makers interested in understanding better how different correspondent banking configurations are related with specific portfolios of services. Researchers interested in the subject of branchless banking can also benefit from the taxonomy presented and the transaction costs analysis of this kind of banking channel, which has been adopted in a number of developing countries all over the world now. (C) 2011 Elsevier B.V. All rights reserved.

On Equilibrium Distribution of a Reversible Growth Model

We study a probabilistic model of interacting spins indexed by elements of a finite subset of the d-dimensional integer lattice, da parts per thousand yen1. Conditions of time reversibility are examined. It is shown that the model equilibrium distribution converges to a limit distribution as the indexing set expands to the whole lattice. The occupied site percolation problem is solved for the limit distribution. Two models with similar dynamics are also discussed.

A transmission problem for beams on nonlinear supports

A transmission problem involving two Euler-Bernoulli equations modeling the vibrations of a composite beam is studied. Assuming that the beam is clamped at one extremity, and resting on an elastic bearing at the other extremity, the existence of a unique global solution and decay rates of the energy are obtained by adding just one damping device at the end containing the bearing mechanism.

Predicting the connectivity of primate cortical networks from topological and spatial node properties

Abstract Background The organization of the connectivity between mammalian cortical areas has become a major subject of study, because of its important role in scaffolding the macroscopic aspects of animal behavior and intelligence. In this study we present a computational reconstruction approach to the problem of network organization, by considering the topological and spatial features of each area in the primate cerebral cortex as subsidy for the reconstruction of the global cortical network connectivity. Starting with all areas being disconnected, pairs of areas with similar sets of features are linked together, in an attempt to recover the original network structure. Results Inferring primate cortical connectivity from the properties of the nodes, remarkably good reconstructions of the global network organization could be obtained, with the topological features allowing slightly superior accuracy to the spatial ones. Analogous reconstruction attempts for the C. elegans neuronal network resulted in substantially poorer recovery, indicating that cortical area interconnections are relatively stronger related to the considered topological and spatial properties than neuronal projections in the nematode. Conclusion The close relationship between area-based features and global connectivity may hint on developmental rules and constraints for cortical networks. Particularly, differences between the predictions from topological and spatial properties, together with the poorer recovery resulting from spatial properties, indicate that the organization of cortical networks is not entirely determined by spatial constraints.

Constraint-based analysis of gene interactions using restricted boolean networks and time-series data

Abstract Background A popular model for gene regulatory networks is the Boolean network model. In this paper, we propose an algorithm to perform an analysis of gene regulatory interactions using the Boolean network model and time-series data. Actually, the Boolean network is restricted in the sense that only a subset of all possible Boolean functions are considered. We explore some mathematical properties of the restricted Boolean networks in order to avoid the full search approach. The problem is modeled as a Constraint Satisfaction Problem (CSP) and CSP techniques are used to solve it. Results We applied the proposed algorithm in two data sets. First, we used an artificial dataset obtained from a model for the budding yeast cell cycle. The second data set is derived from experiments performed using HeLa cells. The results show that some interactions can be fully or, at least, partially determined under the Boolean model considered. Conclusions The algorithm proposed can be used as a first step for detection of gene/protein interactions. It is able to infer gene relationships from time-series data of gene expression, and this inference process can be aided by a priori knowledge available.

Modularity and evolutionary constraints in a baculovirus gene regulatory network

Abstract Background The structure of regulatory networks remains an open question in our understanding of complex biological systems. Interactions during complete viral life cycles present unique opportunities to understand how host-parasite network take shape and behave. The Anticarsia gemmatalis multiple nucleopolyhedrovirus (AgMNPV) is a large double-stranded DNA virus, whose genome may encode for 152 open reading frames (ORFs). Here we present the analysis of the ordered cascade of the AgMNPV gene expression. Results We observed an earlier onset of the expression than previously reported for other baculoviruses, especially for genes involved in DNA replication. Most ORFs were expressed at higher levels in a more permissive host cell line. Genes with more than one copy in the genome had distinct expression profiles, which could indicate the acquisition of new functionalities. The transcription gene regulatory network (GRN) for 149 ORFs had a modular topology comprising five communities of highly interconnected nodes that separated key genes that are functionally related on different communities, possibly maximizing redundancy and GRN robustness by compartmentalization of important functions. Core conserved functions showed expression synchronicity, distinct GRN features and significantly less genetic diversity, consistent with evolutionary constraints imposed in key elements of biological systems. This reduced genetic diversity also had a positive correlation with the importance of the gene in our estimated GRN, supporting a relationship between phylogenetic data of baculovirus genes and network features inferred from expression data. We also observed that gene arrangement in overlapping transcripts was conserved among related baculoviruses, suggesting a principle of genome organization. Conclusions Albeit with a reduced number of nodes (149), the AgMNPV GRN had a topology and key characteristics similar to those observed in complex cellular organisms, which indicates that modularity may be a general feature of biological gene regulatory networks.

Different approaches for dealing with rare variants in family-based genetic studies: application of a Genetic Analysis Workshop 17 problem

Rare variants are becoming the new candidates in the search for genetic variants that predispose individuals to a phenotype of interest. Their low prevalence in a population requires the development of dedicated detection and analytical methods. A family-based approach could greatly enhance their detection and interpretation because rare variants are nearly family specific. In this report, we test several distinct approaches for analyzing the information provided by rare and common variants and how they can be effectively used to pinpoint putative candidate genes for follow-up studies. The analyses were performed on the mini-exome data set provided by Genetic Analysis Workshop 17. Eight approaches were tested, four using the trait’s heritability estimates and four using QTDT models. These methods had their sensitivity, specificity, and positive and negative predictive values compared in light of the simulation parameters. Our results highlight important limitations of current methods to deal with rare and common variants, all methods presented a reduced specificity and, consequently, prone to false positive associations. Methods analyzing common variants information showed an enhanced sensibility when compared to rare variants methods. Furthermore, our limited knowledge of the use of biological databases for gene annotations, possibly for use as covariates in regression models, imposes a barrier to further research.

The Brazilian Network for HIV-1 Genotyping External Quality Control Assurance Programme

The Brazilian network for genotyping is composed of 21 laboratories that perform and analyze genotyping tests for all HIV-infected patients within the public system, performing approximately 25,000 tests per year. We assessed the interlaboratory and intralaboratory reproducibility of genotyping systems by creating and implementing a local external quality control evaluation. Plasma samples from HIV-1-infected individuals (with low and intermediate viral loads) or RNA viral constructs with specific mutations were used. This evaluation included analyses of sensitivity and specificity of the tests based on qualitative and quantitative criteria, which scored laboratory performance on a 100-point system. Five evaluations were performed from 2003 to 2008, with 64% of laboratories scoring over 80 points in 2003, 81% doing so in 2005, 56% in 2006, 91% in 2007, and 90% in 2008 (Kruskal-Wallis, p = 0.003). Increased performance was aided by retraining laboratories that had specific deficiencies. The results emphasize the importance of investing in laboratory training and interpretation of DNA sequencing results, especially in developing countries where public (or scarce) resources are used to manage the AIDS epidemic.

Mapping virtual networks onto substrate networks

Network virtualization is a promising technique for building the Internet of the future since it enables the low cost introduction of new features into network elements. An open issue in such virtualization is how to effect an efficient mapping of virtual network elements onto those of the existing physical network, also called the substrate network. Mapping is an NP-hard problem and existing solutions ignore various real network characteristics in order to solve the problem in a reasonable time frame. This paper introduces new algorithms to solve this problem based on 0–1 integer linear programming, algorithms based on a whole new set of network parameters not taken into account by previous proposals. Approximative algorithms proposed here allow the mapping of virtual networks on large network substrates. Simulation experiments give evidence of the efficiency of the proposed algorithms.

Variance Reduction in Analytical Chemistry : New Numerical Methods in Chemometrics and Molecular Simulation

This thesis is based on five papers addressing variance reduction in different ways. The papers have in common that they all present new numerical methods. Paper I investigates quantitative structure-retention relationships from an image processing perspective, using an artificial neural network to preprocess three-dimensional structural descriptions of the studied steroid molecules. Paper II presents a new method for computing free energies. Free energy is the quantity that determines chemical equilibria and partition coefficients. The proposed method may be used for estimating, e.g., chromatographic retention without performing experiments. Two papers (III and IV) deal with correcting deviations from bilinearity by so-called peak alignment. Bilinearity is a theoretical assumption about the distribution of instrumental data that is often violated by measured data. Deviations from bilinearity lead to increased variance, both in the data and in inferences from the data, unless invariance to the deviations is built into the model, e.g., by the use of the method proposed in paper III and extended in paper IV. Paper V addresses a generic problem in classification; namely, how to measure the goodness of different data representations, so that the best classifier may be constructed. Variance reduction is one of the pillars on which analytical chemistry rests. This thesis considers two aspects on variance reduction: before and after experiments are performed. Before experimenting, theoretical predictions of experimental outcomes may be used to direct which experiments to perform, and how to perform them (papers I and II). After experiments are performed, the variance of inferences from the measured data are affected by the method of data analysis (papers III-V).