Carlos Alberto Brayner de Oliveira Lira

Recent advances in perturbative methods applied to nuclear engineering problems

Abstract This paper describes the main perturbative methods used in sensitivity analysis. It also presents several applications to calculate sensitivity coefficients of interest to nuclear engineering problems. These topics will cover only the papers presented in the 10th Brazilian Meeting on Reactor Physics and Thermal Hydraulics - X ENFIR. First the matrix formalism of the perturbation theory is applied in a simplified model to study the hot channel of PWR reactors. The mass, linear momentum and energy conservation equations and appropriate heat transfer and fluid mechanic correlations describe the discretized system. After calculating systems thermal hydraulic properties, the matrix formalism is applied and the sensitivity coefficients are determined for each case of interest. Comparisons between perturbative and direct calculations show good agreement, which demonstrates that the matrix formalism is an important tool for discretized system analysis. A second application of the GPT methodology to a reliability engineering problem of great practical interest follows: that of the analysis of the influence of the demand rate on the reliability of a process plant equipped with a single protective channel. The application of the GPT approach needs the solution of the system for a few points (reference solutions). The results agree very well with those published in the literature. In a third application, the so called differential method was applied to the sensitivity analysis of the waterhammer effect in hydraulic networks. As an example, a constant-level tank connected through a pipe to a valve discharging to atmosphere was considered. The sensitivity coefficients for given responses obtained by using both the differential method and the response surface generated by the computer code WHAT were calculated. The results obtained show excellent agreement. As a fourth application, the GPT formalism is used to analyze the sensitivity of some responses of interest to boron neutron capture therapy. Attention is focused on diffusive source-driven systems. Finally a methodology is described relevant to the use of data from experimental facilities with respect to a reference design. The method makes use of GPT methods for the calculation of the sensitivity coefficients relevant to the responses of interest with respect to the system parameters. Adjustment and data transposition methods extensively used in the reactor physics neutronic domain are considered as applicable to other fields of interest, in particular to the thermal-hydraulic field.

Improving Disturbance Classification by Combining Multiple Artificial Neural Networks

An ANN-based automatic classifier for power system disturbance waveforms was developed. Actual voltage waveforms were applied in the training process. Signals are processed in two steps: i) decomposition through wavelet transformation up to the 5th decomposition level; ii) the resultant wavelet coefficients are processed via PCA, reducing the input space of the classifier to a much lower dimension. The classification is carried out using a combination of 3 MLPs with different architectures. The RPROP algorithm is applied for training the networks. Network combination was formed and the final decision of the classifier corresponds to the combination output with the highest value. The results showed to be quite promising for five disturbance types tested so far: sags, swells, harmonics, oscillatory transients and interruptions, as well as in the particular case of no disturbance.

Combining Artificial Neural Network for diagnosing polluted insulators

This paper presents a method to classify the current polluted level on insulator surfaces, i.e., to diagnose the operational conditions of the electrical system isolation through pattern recognition techniques using the ultrasonic signals obtained from surface discharges on outdoor insulators. Pattern extraction techniques on the input signals by Artificial Neural Networks were used in order to enable a reliable computation during the training. It can be point out that the area centroid of the ultrasonic signals showed a powerful extraction technique. Here, the Multilayer Perceptron Network was used as a single classifier or as a combination of multiple classifiers. Moreover, the developed networks have one or six neurons in their output layer to represent the classes of pollution. A comparison among the four developed neural net models shows the improvement of the networks with six output neurons and that the use of combined models is a powerful technique for this type of application.

Effects of some calculation parameters on the computational modelling of temperature, velocity and gas volume fraction during steady-state operation of an aqueous homogeneous reactor

This paper is part of the ongoing efforts to contribute to the thermal-hydraulic analysis of one of the most promising alternatives to produce medical isotopes and meeting current and future demand for 99Mo: the use of Aqueous Homogeneous Reactors (AHRs). In this paper, the effects of some calculation parameters like mesh refinement, time step size, turbulence models, transient schemes and numerical advection scheme on the computational modelling of key parameters of an AHR steady-state operation have been investigated. For this purpose, a 75 kWth AHR conceptual design based on the ARGUS reactor, six meshes, five time step sizes, three different models for solving flow problems, three numerical advection schemes and the available transient schemes were used in the simulations. The numerical simulations were carried out using the Computational Fluid Dynamic (CFD) code ANSYS CFX 14. The results of the CFD simulations allow developing a detailed and improved CFD model of the AHR core on which the effects of the investigated calculation parameters are quantifiable.

Efficiency of a conceptual design of a high temperature electrolysis system coupled to a VHTR for nuclear hydrogen production

High temperature electrolysis process coupled to a very high temperature reactor is one of the most promising methods for hydrogen production using a nuclear reactor as the primary energy source. A computational fluid dynamic model for the evaluation and optimisation of the electrolyser of a high temperature electrolysis hydrogen production process flowsheet was developed using ANSYS FLUENT®. Electrolysers operational and design parameters will be optimised in order to obtain the maximum hydrogen production and the higher efficiency in the module. A complete flowsheet is proposed for the high temperature electrolysis process coupled to an accelerator driven system, considering a Brayton cycle for the energy production and a sweep gas system for the gas separation. An acceptable value of global efficiency for the initial operating condition is obtained. Several parametric studies are conducted using the flowsheet proposed to evaluate important operating parameters in the overall process efficiency.

Thermal-Hydraulics Study of a 75 kWth Aqueous Homogeneous Reactor for 99Mo Production

is a very useful radioisotope, which is used in nearly 80% of all nuclear medicine procedures. is produced from 99Mo decay. A potentially advantageous alternative to meeting current and future demand for 99Mo is the use of Aqueous Homogeneous Reactors (AHR). In this paper, a thermal-hydraulics study of the core of a 75 kWth AHR conceptual design based on the ARGUS reactor for 99Mo production is presented. As the ARGUS heat removal systems were designed for working at 20 kWth, the main objective of the thermal-hydraulics study was evaluating the heat removal systems in order to show that sufficient cooling capacity exists to prevent fuel solution overheating. The numerical simulations of an AHR model were carried out using the Computational Fluid Dynamic (CFD) code ANSYS CFX 14. Evaluation shows that the ARGUS heat removal systems working at 75 kWth are not able to provide sufficient cooling capacity to prevent fuel solution overheating. To solve this problem, the number of coiled cooling pipes inside the core was increased from one to five. The results of the CFD simulations with this modification in the design show that acceptable temperature distributions can be obtained.

Daytime Net Radiation on Cowpea and Castor Bean in the County of Areia-PB and Garanhuns-PE

continuum, such as evapotranspiration, however, its direct measurement is not available for most weather stations in Brazil. Thus, the objective of this work was to determine the relationship between Rn and incident solar radiation (Rg) in cowpea and castor bean canopies. The Rn and Rg data were collected from experiments conducted during in the years 2002 and 2003 (cowpea), 2004 and 2005 (castor bean) in Areia, PB (6 o 58’ S, 35 o 41’ W e 620 m), and 2009 (castor bean) in Garanhuns-PE (8 o 53’ S, 36 o 29’ W e 842 m). Were performed daily sums of Rn and Rg, from 6 to 18h, and then it was calculated the Rn/Rg ratio for each day. Models with high R 2 (above 0.93), d and c indices (above 0.90) and low RMSE (below 0.83) were obtained in test between measured and estimated values, indicating precision to estimate Rn in cowpea and castor bean canopies as a function of Rg. The general linear function for castor bean and cowpea were Rn = 0.5965 Rg (R 2 = 0.943) and Rn = 0.7822 Rg (R 2 = 0.976), respectively. Net radiation can be satisfactorily estimated from incoming solar radiation for both crops.

Fluxos de água e de energia em feijão-caupi em condições de sequeiro, no Nordeste Brasileiro - DOI:10.5039/agraria.v8i2a2090

Em areas de agricultura de sequeiro a quantificacao das perdas de agua por meio da evapotranspiracao (ET), assume grande importância. Os metodos dos balancos de agua (BH) e de energia (BE) vem sendo muito utilizados devido a sua simplicidade, robustez e menor custo. O presente trabalho objetivou avaliar os componentes do BH e do BE em solo cultivado com feijao caupi alem de se fazer comparacoes entre os metodos de determinacao da ET; para tal foram realizados experimentos numa area de 4 ha em Areia, PB (6 o 58’ S, 35 o 41’ W); para determinacao do BE a area foi instrumentada com um pluviografo, um piranometro, um radiometro e sensores para medidas da temperatura e da umidade relativa do ar em dois niveis. Com vista a determinacao do BH, foram instalados tres sitios tensio-neutronicos, contendo um tubo de acesso para sonda de neutrons e oito tensiometros. O valor medio da razao Rn/Rg foi de 76% sendo utilizado, em media, como 73% no fluxo de calor latente, 18% como fluxo de calor sensivel e 9% como fluxo de calor no solo. Verificou-se que o BH e o BE podem ser usados para determinacao da ET. Os valores medios da ET foram de 4,20 e 4,28 mm dia -1 , determinados pelos metodos BH e BE, respectivamente.

Study of Thorium Fuel Cycles for Light Water Reactor VBER-150

The main objective of this paper is to examine the use of thorium-based fuel cycle for the transportable reactors or transportable nuclear power plants (TNPP) VBER-150 concept, in particular the neutronic behavior. The thorium-based fuel cycles included Th232

Boosting algorithm to improve a voltage waveform classifier based on artificial neural network

An ANN-based classifier for voltage wave disturbance was developed. Voltage signals captured on the power transmission system of CHESF, Federal Power Utility, were processed in two steps: by wavelet transform and principal component analysis. The classification was carried out using a combination of six MLPs with different architectures: five representing the first to fifth-level details, and one representing the fifth-level approximation. Network combination was formed using the boosting algorithm which weights a models contribution by its performance rather than giving equal weight to all models. Experimental results with real data indicate that boosting is clearly an effective way to improve disturbance classification accuracy when compared with the simple average and the individual models.