ROBERTO NAVARRO DE MESQUITA

ROBERTO NAVARRO DE MESQUITA

(Fonte: Lattes)

Resumo

Possui graduação em Física pela Universidade Estadual de Campinas (1987), mestrado em Física pela Universidade Estadual de Campinas (1991) e doutorado em Engenharia Mecânica pela Universidade de São Paulo (2002). Atualmente é tecnologista pleno da Comissão Nacional de Energia Nuclear. Tem experiência na área de Ciência da Computação, com ênfase em Sistemas de Inteligência Artificial, atuando principalmente nos seguintes temas: inteligência artificial, diagnóstico de defeitos em tubos, correntes parasitas (ECT), reconhecimento de padrões em imagens. (Texto extraído do Currículo Lattes em 27 dez. 2021)

Página do item completo

Resultados de Busca

Agora exibindo 1 - 10 de 59

Módulo de saída do sistema computacional CaReR
2023 - BELCHIOR JUNIOR, ANTONIO; RIBEIRO, MARIA A.M.; MESQUITA, ROBERTO N. de
Este trabalho tem como objetivo detalhar o Módulo de Saída do Sistema Computacional de Avaliação do Ativo Isotópico. Este sistema desempenha a função de avaliar o inventário radioisotópico dos rejeitos radioativos gerados na CNAAA (Central Nuclear Almirante Álvaro Alberto). Além disso, ele é responsável por estimar o inventário radioisotópico básico do depósito inicial da CNAAA, onde estão armazenados rejeitos radioativos com atividades baixas e intermediárias, com base nos dados da geração de rejeitos radioativos provenientes das usinas nucleoelétricas da CNAAA, dados esses obtidos do RejAn. O módulo de Saída do programa, em conjunto com os módulos de Cálculo e de Entrada, constitui o Programa CaReR. O módulo de Entrada permite ao usuário atualizar as bibliotecas de dados utilizadas pelo programa, seja importando dados do RejAn, da tabela de Fatores de Escala (FE) e do arquivo de Dados de Radionuclídeos e Cadeias de Decaimento. Detalhes sobre o módulo de Entrada foram previamente descritos na referência [1] e as alterações deste módulo que ocorreram durante o desenvolvimento do programa são oportunamente apresentadas neste relatório. Por outro lado, o módulo de Saída, objetivo principal deste relatório, permite ao usuário gerar relatórios, fichas de dados e etiquetas para o controle dos embalados de rejeitos, especialmente quando estes precisam ser movidos para o depósito definitivo. Além disso, este módulo inclui funcionalidades que permitem um acompanhamento detalhado da aplicação da metodologia de cálculo incluindo aplicação de Fatores de Escala e utilização das cadeias de decaimento. Neste documento são ainda fornecidas informações sobre a arquitetura e o diagrama esquemático do software desenvolvido. Este documento foi elaborado para atender o item 8.4.5 do Plano de Trabalho, que inclui a "Especificação do módulo de saída do Programa de Avaliação do Ativo Isotópico incluindo a descrição da arquitetura e diagrama esquemático lógico do software" do Acordo de Pesquisa e Desenvolvimento (Ativo Isotópico - Fase B), celebrado entre a ETN e o IPEN.
CFD Simulation of isothermal upward two-phase flow in a vertical annulus using interfacial area transport equation
2023 - CERAVOLO, FLAVIO E.; ROCHA, MARCELO da S.; MESQUITA, ROBERTO N. de; ANDRADE, DELVONEI A. de
This work presents a numerical simulation of a vertical, upward, isothermal two-phase flow of air bubbles and water in an annular channel applying a Computational Fluid Dynamics (CFD) code. For this, the Two-Fluid model is applied considering interfacial force correlations, namely: drag, lift, wall lubrication, turbulent dispersion, and virtual mass. The turbulence k-ε model effects and the influence of One-group Interfacial Area Transport Equation (IATE) are taken into account, in this case, the influence of two source term correlations for the bubble breakup and coalescence IATE is analysed. The work assesses whether the code properly represents the physical phenomenon by comparing the simulation results with experimental data obtained from the literature. Six flow conditions are evaluated based on two superficial liquid velocities and three void fractions in the bubbly flow regimen. The annular channel adopted has an outer pipe with an internal diameter of 38.1 mm and an inner cylinder of 19.1 mm. To represent this geometry, a three-dimensional mesh was generated with 160,000 elements, after a mesh sensitivity study. The void fraction distribution, taken radially to the flow section, is the main parameter analysed as well as interfacial area concentration, interfacial gas velocity, and bubble sizes distribution. The CFD model implemented in this work demonstrates satisfactory agreement with the reference experimental data but indicates the need for further improvement in the phase interaction models.
Critical velocity experimental assessment in flat plate fuel element for nuclear research reactor
2022 - ANDRADE, D.A.; MANTECON, J.G.; MESQUITA, R.N.; MATTAR NETO, M.; UMBEHAUN, P.E.; TORRES, W.M.
Aluminum-coated plates, containing a uranium silicide (U3Si2) meat dispersed in an aluminum matrix, are commonly used in the fuel elements of Material Testing Reactors (MTRs). These fuel elements are typically comprised of narrow channels formed by parallel flat plates, which allow coolant flow to remove the heat of fission reactions. It is important to mention that the thickness of the plates is much smaller than their width and height. The high flow rates needed to ensure efficient fuel-element cooling may cause fuel-plate mechanical failures due to instability induced by the flow in the channels. In the case of critical velocity, excessive permanent deflections of these plates can cause blockage of the flow channels and lead to overheating. An experimental facility that simulates a plate-like fuel element with three coolant channels was developed for this work. The test-section dimensions were based on the Fuel Element design of the Brazilian Multipurpose Reactor (RMB), project being coordinated by the National Commission of Nuclear Energy (CNEN). Experiments were performed to reach Miller's critical velocity condition. This critical condition was reached at 14.5 m/s leading to consequent plastic deformation of the fuel plates.
A comparative study on machine learning regression algorithms aplied to modeling gas centrifuge
2022 - ANDRADE, DELVONEI A. de; MESQUITA, ROBERTO N. de; NASCIMENTO, NATAN P.
The gas Centrifuge is a very hard equipment to model, because it involves a gas dynamic with many complications, such as hypersonic waves and rarefied regions combined with continuous flow areas. Therefore, data analysis regressions remain currently a very important technique to understand and describe the problem in a practical way. This paper intends to apply and compare several regression techniques using machine learning, to obtain a hydraulic and a separative power model of gas centrifuge used in enrichment plants. For this purpose, a set of normalized data composed of 134 experimental lines was used, observing the variables of interest, the separation power (dU), and the waste pressure (Pw), through the following explanatory variables: feed flow (F), cut (q), and product pressure (Pp). The comparisons were presented between the results obtained for the models generated by the following: algorithms, multivariate regression, multivariate adaptive regression splines – MARS, bootstrap aggregating multivariate adaptive regression splines – Bagging MARS, artificial neural network – ANN, extreme gradient boosting – XGBoost, support vector regression– Poly SVR, radial basis Function support vector regression – RBF SVR, K-nearest neighbors – KNN and Stacked Ensemble. That way, to avoid overfitting and provide insights about generalization of the models in unseen data, during the training phase, the k-fold cross validation approach was used. Subsequently, the residuals were analyzed, and the models were compared by the following metrics: Root mean square error – RMSE; Mean squared error – MSE; Mean absolute error – MAE; and Coefficient of determination – R2.
Prediction of failures in rotating machines by vibration spectral analysis
2021 - POVEDA, PEDRO F.; MESQUITA, ROBERTO N. de
Classification of two-phase flow instability phases using convolutional neural networks
2021 - SCHOTT, S.M.C.; SILVA, M.C.B. da; MESQUITA, R.N. de
Characterization of polyacrylonitrile thermal stabilization process for carbon fiber production using intelligent algorithms
2021 - TERRA, BRUNA M.; ANDRADE, DELVONEI A. de; MESQUITA, ROBERTO N. de
Composite materials have widened their application range in recent years. The polymeric composite reinforced with carbon fibers can be described as a high-performance structural material which merges two important features: low weight and mechanical stability. Carbon fiber production which uses polyacrylonitrile as precursor is composed of many stages such as polymerization, spinning, thermal stabilization, carbonization, and surface treatment. Thermal stabilization is the critical stage of this production process, during which aromatic rings are generated, and therefore the main factor for the carbon fiber structure definition and thus for this material quality. A thermal stabilization model using intelligent algorithms was developed aiming a possible optimization of the production process and consequent cost reduction. This work was based on real experimental data obtained from a composite material production pilot plant. A qualitative analysis was initially performed using Self-Organizing Maps trained with variables of fiber production reagents and process. Thereafter, a supervised training with feedforward backpropagation neural network was used for a quantitative analysis. Based on this quantitative analysis, the carbon fiber thermal stabilization process was simulated, obtaining 2.98% and 2.48% mean errors relative to experimental results of Volumetric Density and FTIR Conversion index, respectively.
Nutritional evaluation of Brachiaria brizantha cv. marandu using convolutional neural networks
2020 - DAL PRÁ, BRUNO R.; MESQUITA, ROBERTO N. de; MENEZES, MARIO O. de; ANDRADE, DELVONEI A. de
The identification of plant nutritional stress based on visual symptoms is predominantly done manually and is performed by trained specialists to identify such anomalies. In addition, this process tends to be very time consuming, has a variability between crop areas and is often required for analysis at various points of the property. This work proposes an image recognition system that analyzes the nutritional status of the plant to help solve these problems. The methodology uses deep learning that automates the process of identifying and classifying nutritional stress of Brachiaria brizantha cv. marandu. An image recognition system was built and analyzes the nutritional status of the plant using the digital images of its leaves. The system identifies and classifies Nitrogen and Potassium deficiencies. Upon receiving the image of the pasture leaf, after a classification performed by a convolutional neural network (CNN), the system presents the result of the diagnosed nutritional status. Tests performed to identify the nutritional status of the leaves presented an accuracy of 96%. We are working to expand the data of the image database to obtain an increase in the accuracy levels, aiming at the training with a larger amount of information presented to CNN and, thus, obtaining results that are more expressive.
Development of a real-time focus estimaton software to be applied in two-phase flow imaging using intelligent processing
2019 - SILVA, MARCONES C.B. da; SCHOTT, SANDRO M.C.; MESQUITA, ROBERTO N. de
Image processing has been an increasing research area in the last decades, especially due to crescent technological growth allied with lowering production costs. Many scientific applications have searched for establishment of quality norms associated with possible information obtainment from images. A common need from different applications has been the standardization of focus quality metric. The development of new methods for measuring the focus adjustment in order to obtain image quality metric analysis has enabled more reliable and precise data in many different industry and science sectors. Some examples are industrial equipment parts inspection using computational vision to defects classification. This work presents the initial steps to develop a methodology to estimate focus in real time in two-phase flow experiments inside tube with cylindrical geometry. This methodology is initially based on a software module using artificial intelligence methods to estimate image focus. This module is developed in LabVIEW platform using Fuzzy Logic inference base in different traditional digital focus metrics and integrated with digital cameras to increment precision on focus adjustment during two-phase flow experiments. This method will be calibrated to be used on void fraction estimation through image analysis in the natural circulation loop located at the Nuclear Engineering Center (CEN) do Instituto de Pesquisas Energéticas e Nucleares (IPEN). A set of the initial developed software modules will be presented with their respective functionalities, initial results and experimental focus estimated errors.
Two-phase flow void fraction estimation based on bubble image segmentation using Randomized Hough Transform with Neural Network (RHTN)
2020 - SERRA, PEDRO L.S.; MASOTTI, PAULO H.F.; ROCHA, MARCELO S.; ANDRADE, DELVONEI A. de; TORRES, WALMIR M.; MESQUITA, ROBERTO N. de
The International Atomic Energy Agency (IAEA) has been encouraging the use of passive cooling systems in new designs of nuclear power plants. Next nuclear reactor generations are intended to have simpler and robust safety resources. Natural Circulation based systems hold an undoubtedly prominent position among these. The study of limiting conditions of these systems has led to instability behavior analysis where many different two-phase flow patterns are present. Void fraction is a key parameter in thermal transfer analysis of these flow instability conditions. This work presents a new method to estimate void fraction from images captured of an experimental two-phase flow circuit. The method integrates a set of Artificial Neural Networks with a modified Randomized Hough Transform to make multiple scans over acquired images, using crescent-sized masks. This method was called Randomized Hough Transform with Neural Network (RHTN). Each different mask size is chosen according with bubble sizes, which are the main ‘objects of interest’ in this image analysis. Images are segmented using fuzzy inference with different parameters adjusted based on acquisition focus. Void fraction calculation considers the volume of the imaged geometrical section of flow inside cylindrical glass tubes considering the acquisition depth-of-field used. The bubble volume is estimated based on geometrical parameters inferred for each detected bubble. The image database is obtained from experiments performed on a vertical two-phase flow circuit made of cylindrical glass where flow-patterns visualization is possible. The results have shown that the estimation method had good agreement with increasing void fraction experimental values. RHTN has been very efficient as bubble detector with very low ‘false-positive’ cases (< 0.004%) due robustness obtained through integration between Artificial Neural Networks with Randomized Hough Transforms.