Optimization of Radial Basis Function neural network employed for prediction of surface roughness in hard turning process using Taguchi's orthogonal arrays

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Local dimension and finite time prediction in coupled map lattices

Forecasting, for obvious reasons, often become the most important goal to be achieved. For spatially extended systems (e.g. atmospheric system) where the local nonlinearities lead to the most unpredictable chaotic evolution, it is highly desirable to have a simple diagnostic tool to identify regions of predictable behaviour. In this paper, we discuss the use of the bred vector (BV) dimension, a recently introduced statistics, to identify the regimes where a finite time forecast is feasible. Using the tools from dynamical systems theory and Bayesian modelling, we show the finite time predictability in two-dimensional coupled map lattices in the regions of low BV dimension. © Indian Academy of Sciences.

Neural network approach for surface roughness prediction in surface grinding

Several systems are currently tested in order to obtain a feasible and safe method for automation and control of grinding process. This work aims to predict the surface roughness of the parts of SAE 1020 steel ground in a surface grinding machine. Acoustic emission and electrical power signals were acquired by a commercial data acquisition system. The former from a fixed sensor placed near the workpiece and the latter from the electric induction motor that drives the grinding wheel. Both signals were digitally processed through known statistics, which with the depth of cut composed three data sets implemented to the artificial neural networks. The neural network through its mathematical logical system interpreted the signals and successful predicted the workpiece roughness. The results from the neural networks were compared to the roughness values taken from the worpieces, showing high efficiency and applicability on monitoring and controlling the grinding process. Also, a comparison among the three data sets was carried out.

Risk prediction for weed infestation using classification rules

This paper proposes a fuzzy classification system for the risk of infestation by weeds in agricultural zones considering the variability of weeds. The inputs of the system are features of the infestation extracted from estimated maps by kriging for the weed seed production and weed coverage, and from the competitiveness, inferred from narrow and broad-leaved weeds. Furthermore, a Bayesian network classifier is used to extract rules from data which are compared to the fuzzy rule set obtained on the base of specialist knowledge. Results for the risk inference in a maize crop field are presented and evaluated by the estimated yield loss. © 2009 IEEE.

Cluster analysis and artificial neural network on the superovulatory response prediction in mice

Complex biological systems require sophisticated approach for analysis, once there are variables with distinct measure levels to be analyzed at the same time in them. The mouse assisted reproduction, e.g. superovulation and viable embryos production, demand a multidisciplinary control of the environment, endocrinologic and physiologic status of the animals, of the stressing factors and the conditions which are favorable to their copulation and subsequently oocyte fertilization. In the past, analyses with a simplified approach of these variables were not well succeeded to predict the situations that viable embryos were obtained in mice. Thereby, we suggest a more complex approach with association of the Cluster Analysis and the Artificial Neural Network to predict embryo production in superovulated mice. A robust prediction could avoid the useless death of animals and would allow an ethic management of them in experiments requiring mouse embryo.

On the influence of the mode-shapes of a marine propulsion shafting system on the prediction of torsional stresses

Torsional vibration predictions and measurements of a marine propulsion system, which has both damping and a highly flexible coupling, are presented in this paper. Using the conventional approach to stress prediction in the shafting system, the numerical predictions and the experimental torsional vibration stress curves in some parts of the shafting system are found to be quite different. The free torsional vibration characteristics and forced torsional vibration response of the system are analyzed in detail to investigate this phenomenon. It is found that the second to fourth natural modes of the shafting system have significant local deformation. This results in large torsional resonant responses in different sections of the system corresponding to different engine speeds. The results show that when there is significant local deformation in the shafting system for different modes, then multi-point measurements should be made, rather than the conventional method of using a single measurement at the free end of the shaft, to obtain the full torsional vibration characteristics of the shafting system.

Prediction of dopant atom distribution on nanocrystals using thermodynamic arguments

A theoretical approach aiming at the prediction of segregation of dopant atoms on nanocrystalline systems is discussed here. It considers the free energy minimization argument in order to provide the most likely dopant distribution as a function of the total doping level. For this, it requires as input (i) a fixed polyhedral geometry with defined facets, and (ii) a set of functions that describe the surface energy as a function of dopant content for different crystallographic planes. Two Sb-doped SnO2 nanocrystalline systems with different morphology and dopant content were selected as a case study, and the calculation of the dopant distributions expected for them is presented in detail. The obtained results were compared to previously reported characterization of this system by a combination of HRTEM and surface energy calculations, and both methods are shown to be equivalent. Considering its application pre-requisites, the present theoretical approach can provide a first estimation of doping atom distribution for a wide range of nanocrystalline systems. We expect that its use will support the reduction of experimental effort for the characterization of doped nanocrystals, and also provide a solution to the characterization of systems where even state-of-art analytical techniques are limited.

Prediction of Oncogenic Interactions and Cancer-Related Signaling Networks Based on Network Topology

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

An evolutionary approach to improve connectivity prediction in mobile wireless sensor networks

Connectivity is the basic factor for the proper operation of any wireless network. In a mobile wireless sensor network it is a challenge for applications and protocols to deal with connectivity problems, as links might get up and down frequently. In these scenarios, having knowledge of the node remaining connectivity time could both improve the performance of the protocols (e.g. handoff mechanisms) and save possible scarce nodes resources (CPU, bandwidth, and energy) by preventing unfruitful transmissions. The current paper provides a solution called Genetic Machine Learning Algorithm (GMLA) to forecast the remainder connectivity time in mobile environments. It consists in combining Classifier Systems with a Markov chain model of the RF link quality. The main advantage of using an evolutionary approach is that the Markov model parameters can be discovered on-the-fly, making it possible to cope with unknown environments and mobility patterns. Simulation results show that the proposal is a very suitable solution, as it overcomes the performance obtained by similar approaches.

Prediction of genomic-enabled breeding values and genome-wide association study for feedlot average daily weight gain in Nelore cattle

Pós-graduação em Genética e Melhoramento Animal - FCAV

Prediction of soil shear strength in agricultural and natural environments of the Brazilian Cerrado

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Transcatheter aortic valve implantation with a self-expanding nitinol bioprosthesis: Prediction of the Need for Permanent Pacemaker Using Simple Baseline and Procedural Characteristics

Objective: To ascertain incidence and predictors of new permanent pacemaker (PPM) following transcatheter aortic valve implantation (TAVI) with the self-expanding aortic bioprosthesis. Background: TAVI with the Medtronic Corevalve (MCV) Revalving System (Medtronic, Minneapolis, MN) has been associated with important post-procedural conduction abnormalities and frequent need for PPM. Methods: Overall, 73 consecutive patients with severe symptomatic AS underwent TAVI with the MCV at two institutions; 10 patients with previous pacemaker and 3 patients with previous aortic valve replacement were excluded for this analysis. Clinical, echocardiographic, and procedural data were collected prospectively in a dedicated database. A standard 12-lead ECG was recorded in all patients at baseline, after the procedure and predischarge. Decision to implant PPM was taken according to current guidelines. Logistic multivariable modeling was applied to identify independent predictors of PPM at discharge. Results: Patients exhibited high-risk features as evidenced by advanced age (mean = 82.1 +/- 6.2 years) and high surgical scores (logistic EuroSCORE 23.0 +/- 12.8%, STS score 9.4 +/- 6.9%). The incidence of new PPM was 28.3%. Interventricular septum thickness and logistic Euroscore were the baseline independent predictors of PPM. When procedural variables were included, the independent predictors of PPM were interventricular septum thickness (OR 0.52; 95% CI 0.320.85) and the distance between noncoronary cusp and the distal edge of the prosthesis (OR 1.37; 95% CI 1.031.83). Conclusions: Conduction abnormalities are frequently observed after TAVI with self-expandable bioprosthesis and definitive pacing is required in about a third of the patients, with a clear association with depth of implant and small interventricular septum thickness. (c) 2011 Wiley Periodicals, Inc.

Liquid-liquid equilibria for systems composed of rice bran oil and alcohol-rich solvents: Application to extraction and deacidification of oil

The liquid-liquid equilibria of systems composed of rice bran oil, free fatty acids, ethanol and water were investigated at temperatures ranging from 10 to 60 degrees C. The results of the present study indicated that the mutual solubility of the compounds decreased with an increase in the water content of the solvent and a decrease in the temperature of the solution. The experimental data set was correlated by applying the UNIQUAC model. The average variance between the experimental and calculated compositions was 0.35%, indicating that the model can accurately predict behavior of the compounds at different temperatures and degrees of hydration. The adjustment of interaction parameters enables both the simulation of liquid-liquid extractors for deacidification of vegetable oil and the prediction of phase compositions for the oil and alcohol-rich phases that are generated during cooling of the stream exiting the extractor (when using ethanol as the solvent). (C) 2012 Elsevier Ltd. All rights reserved.

Identification of novel components of NAD-utilizing metabolic pathways and prediction of their biochemical functions

Nicotinamide adenine dinucleotide (NAD) is a ubiquitous cofactor participating in numerous redox reactions. It is also a substrate for regulatory modifications of proteins and nucleic acids via the addition of ADP-ribose moieties or removal of acyl groups by transfer to ADP-ribose. In this study, we use in-depth sequence, structure and genomic context analysis to uncover new enzymes and substrate-binding proteins in NAD-utilizing metabolic and macromolecular modification systems. We predict that Escherichia coli YbiA and related families of domains from diverse bacteria, eukaryotes, large DNA viruses and single strand RNA viruses are previously unrecognized components of NAD-utilizing pathways that probably operate on ADP-ribose derivatives. Using contextual analysis we show that some of these proteins potentially act in RNA repair, where NAD is used to remove 2'-3' cyclic phosphodiester linkages. Likewise, we predict that another family of YbiA-related enzymes is likely to comprise a novel NAD-dependent ADP-ribosylation system for proteins, in conjunction with a previously unrecognized ADP-ribosyltransferase. A similar ADP-ribosyltransferase is also coupled with MACRO or ADP-ribosylglycohydrolase domain proteins in other related systems, suggesting that all these novel systems are likely to comprise pairs of ADP-ribosylation and ribosylglycohydrolase enzymes analogous to the DraG-DraT system, and a novel group of bacterial polymorphic toxins. We present evidence that some of these coupled ADP-ribosyltransferases/ribosylglycohydrolases are likely to regulate certain restriction modification enzymes in bacteria. The ADP-ribosyltransferases found in these, the bacterial polymorphic toxin and host-directed toxin systems of bacteria such as Waddlia also throw light on the evolution of this fold and the origin of eukaryotic polyADP-ribosyltransferases and NEURL4-like ARTs, which might be involved in centrosomal assembly. We also infer a novel biosynthetic pathway that might be involved in the synthesis of a nicotinate-derived compound in conjunction with an asparagine synthetase and AMPylating peptide ligase. We use the data derived from this analysis to understand the origin and early evolutionary trajectories of key NAD-utilizing enzymes and present targets for future biochemical investigations.