Cary, NC, United States
Cary, NC, United States

Time filter

Source Type

Ozar B.,Fauske and Associates | Harvill R.,Numerical Applications Inc. | Henry C.E.,Fauske and Associates | Norton D.A.,Numerical Applications Inc.
International Conference on Nuclear Engineering, Proceedings, ICONE | Year: 2012

A study to characterize the steam waterhammer phenomena of a low pressure cooling injection (LPCI) system for a Mark 1 boiling water reactor (BWR) has been performed using RELAP5 and GOTHIC during a transient event. The scenario of particular interest was a manual switchover from shutdown cooling mode 3 to low pressure injection due to a loss of coolant accident (LOCA). This transient was initiated by opening the isolation valves of the two trains on a LPCI system into the torus. The torus was considered to be at atmospheric pressure and 20°C. The initial condition of the problem was set up such that the liquid was stagnant in the system. The initial temperature and pressure of the liquid, which was between the torus and isolation valves, was considered to be the same as the torus conditions. On the other hand, the initial condition of the liquid upstream of the isolation valves was chosen to be at 1 MPa and near saturation temperature. The analysis showed that the liquid in the system flashed into steam and discharged into the torus after the isolation valves started to open. Discharge of steam continued until the pressure in the LPCI system reached to a hydrostatic equilibrium with the torus. Following this, the cold liquid from the torus began to reflod the LPCI piping while condensing the steam at the liquid-steam interphase. This caused a mild steam waterhammer event when all of the steam condensed in the piping segments with closed ends. A sensitivity analysis showed that, the magnitude of the steam waterhammer predicted by both codes was sensitive to the number of nodes selected to model the piping, where the steam waterhammer phenomena occurred. Technical basis was obtained from the available literature and used as a guide to choose the number of nodes for the models in both codes. Once the steam waterhammer and the hydrodynamic properties associated with this transient were predicted by both codes, the forces exerted on critical pipe components were calculated. Also, selected thermal-hydraulic properties and hydrodynamic loads were compared between both code calculations. Comparisons showed reasonable agreements. Copyright © 2012 by ASME.


Ozdemir O.E.,Numerical Applications Inc. | George T.L.,Numerical Applications Inc. | Marshall M.D.,Numerical Applications Inc.
Annals of Nuclear Energy | Year: 2015

This paper is a part of Fukushima Technical Evaluation Project (EPRI, 2013a, 2014a, 2015) which investigates various aspects of the Fukushima Daiichi event using the GOTHIC 1 code. The analysis takes advantage of the capability of GOTHIC to model certain aspects of the system geometry and behavior in more detail than typically considered in containment performance analysis. GOTHIC is a general purpose thermal hydraulics code that is used extensively in the nuclear industry for system design support, licensing support and safety analysis. It has the capability to model 3-dimensional flow behavior including the effects of turbulence, diffusion and buoyancy (EPRI, 2014b). This allows GOTHIC to be used in cases where mixing effects and stratification are important. The analysis presented here considers the events at Fukushima Daiichi Unit 1 (1F1) following the tsunami and leading up to the time of the hydrogen detonation in the 1F1 Reactor Building. The 1F1 MAAP5 Baseline Scenario (EPRI, 2013b) is used to define the steam, hydrogen and carbon-monoxide source terms from the primary system and the core concrete interaction. The model incorporates three dimensional modeling of the drywell, wetwell and connecting vent system that can predict the 3-dimensional flow patterns and the temperature and gas distributions. The model also includes leakage to the surrounding reactor building and the wetwell vent to the stack. The 3D containment model includes models for the heat transfer from the steam and gas in the drywell vent system to the torus room, wetwell gas space and pool. Inclusion of vent heat transfer had a significant impact on the overall containment response for the 1F1 scenario, particularly during the steam and hydrogen release from the primary system following the postulated failure of reactor vessel. Condensation in the vent system reduced transfer of noncondensing gases to the wetwell resulting in lower containment pressure and higher gas concentrations in the drywell. © 2015 Elsevier Ltd. All rights reserved.


Ozdemir O.E.,Numerical Applications Inc. | George T.L.,Numerical Applications Inc.
Annals of Nuclear Energy | Year: 2015

As a part of the GOTHIC (GOTHIC incorporates technology developed for the electric power industry under the sponsorship of EPRI.) Fukushima Technical Evaluation project (EPRI, 2014a, b, 2015), GOTHIC (EPRI, 2014c) has been benchmarked against test data for pool stratification (EPRI, 2014a, b, Ozdemir and George, 2013). These tests confirmed GOTHIC's ability to simulate pool mixing and stratification under a variety of anticipated suppression pool operating conditions. The multidimensional modeling requires long simulation times for events that may occur over a period of hours or days. For these scenarios a lumped model of the pressure suppression chamber is desirable to maintain reasonable simulation times. However, a lumped model for the pool is not able to predict the effects of pool stratification that can influence the overall containment response. The main objective of this work is on the development of a correlation that can be used to estimate pool mixing and stratification effects in a lumped modeling approach. A simplified lumped GOTHIC model that includes a two zone model for the suppression pool with controlled circulation between the upper and lower zones was constructed. A pump and associated flow connections are included to provide mixing between the upper and lower pool volumes. Using numerically generated data from a multidimensional GOTHIC model for the suppression pool, a correlation was developed for the mixing rate between the upper and lower pool volumes in a two-zone, lumped model. The mixing rate depends on the pool subcooling, the steam injection rate and the injection depth. © 2015 Elsevier Ltd. All rights reserved.


Andreani M.,Paul Scherrer Institute | Paladino D.,Paul Scherrer Institute | George T.,Numerical Applications Inc.
Nuclear Engineering and Design | Year: 2010

The paper reports the results of the assessment of the GOTHIC code using the data of four basic tests with condensation performed in the PANDA large-scale facility. Three of these experiments featured vertical injection, and in one the transient response due to a high-momentum horizontal injection (jet) was investigated. The injected fluid was either saturated steam or a superheated mixture of steam and helium, and the fluid initially present in the vessels was pure air. The simulations were carried out using a detailed three-dimensional representation of the vessels. In general, the results of the simulations, which used a relatively coarse mesh, were in good agreement with the data. Limitations in modelling local phenomena controlled by complex flow patterns (e.g. heat transfer in the region of an impinging jet) and the need for refined meshes to reproduce certain aspects of the transients (e.g. erosion of the interface between layers of different gas composition) were also identified. Finally, the analyses indicated that in two tests the role played by re-vaporisation of the condensate film was unexpectedly large, and this effect should be more carefully considered in the containment analyses and future model developments. © 2010 Elsevier B.V. All rights reserved.


Ozdemir O.E.,Numerical Applications Inc. | George T.L.,Numerical Applications Inc.
Transactions of the American Nuclear Society | Year: 2013

In this study the GOTHIC model performance on condensation, stratification and mixing phenomena in a pool of water was investigated. GOTHIC models for the POOLEX STB-20 test, with different grid refinements, were compared against the transient temperature distribution in the pool and the vertical temperature profile after 14, 600 seconds of continuous steam injection. Contrary to the experience reported by Li [2] for this test, the integrated model for the pool and discharge pipe gives reasonable results without resorting to special treatment of the heat and momentum source. Increasing the number of cells in the pool generally provided more accurate predictions in comparison to the experimental data. In summary, it is concluded that current GOTHIC models are capable of capturing the steam condensation in a discharge pipe and the thermal stratification inside the water pool correctly for the low steam injection rate of POOLEX test STB-20. Simulations for tests at higher steam injection rate have not yet been completed with the described modeling approach. Consequently, further work is needed to confirm the adequacy of the GOTHIC models for steam condensation and bubble migration in the pool.


Lane J.W.,Numerical Applications Inc. | George T.L.,Numerical Applications Inc.
International Topical Meeting on Nuclear Reactor Thermal Hydraulics 2015, NURETH 2015 | Year: 2015

GOTHIC™ is a versatile and generally applicable software package that solves complex thermal hydraulics problems. GOTHIC solves the conservation equations for mass, momentum and energy for multicomponent, multi-phase flow in lumped parameter and/or multi-dimensional geometries. A distinctive feature of GOTHIC is the ability to track different types of components in the continuous liquid field, including SOLID PARTICLES and DISSOLVED GAS. Each component type provides a unique capability. For example, SOLID PARTICLE components occupy volume within the continuous liquid fields and impact the bulk liquid mass and energy conservation equations (via volume fraction occupied and effective values for the fluid density and heat capacity). As the name implies, a SOLID PARTICLE component is intended to model solid particles, but by using appropriate input parameters, aqueous solutions such as boric acid can be modeled. The convective transport and diffusion is calculated for each component type along with appropriate source terms (e.g., for SOLID PARTICLES these include gravitational settling or buoyancy, resuspension and bed load transport). A component can be included in the initial conditions and added at a boundary condition. Given the different types of components and the fundamental physics included in GOTHIC (e.g., viscous shear and diffusion), the component tracking capability substantially extends its range of applications. Example applications include: boron tracking and mixing in the primary system, effects of gas release on water hammer events, and debris tracking related to GSI-191. Current versions of GOTHIC restrict users to a single liquid component per simulation. The current development effort removes this restriction and allows multiple components (both number and type) to be tracked in a single simulation. This new capability allows: multiple dissolved gases, soluble boron and dissolved gases to co-exist, and multiple debris fields with different characteristic sizes/properties to be considered. In this paper, the assumptions, impact on the other conservation equations, and the solution scheme are discussed for each type of tracked component. Results from simple canonical problems used to verify this new feature are also presented. GOTHIC™ incorporates technology developed for the electric power industry under the sponsorship of EPRI, the Electric Power Research Institute.


Moore T.M.,Numerical Applications Inc. | George T.L.,Numerical Applications Inc.
International Topical Meeting on Nuclear Reactor Thermal Hydraulics 2015, NURETH 2015 | Year: 2015

The Thai (Thermal hydraulics, Aerosols and Iodine) facility, located in Eschborn, Germany, is a 60 m3 steel test vessel designed to simulate operational and accident conditions in a nuclear containment structure. The Thai facility provides experimental data used for validation of thermal hydraulic codes. The test performed at this facility has been modeled using the 1GOTHIC 8.1(QA) software package for the purpose of validating both the physical models and modeling techniques. The test analyzed is from Step 2 of the ISP-47 test performed at the Thai facility. This test consisted of three injection ports for steam and helium to enter the vessel asymmetrically. The asymmetry along with the compartmented geometry of the facility provide a complex coupling of physics that would be present in an accident transient inside of the containment of a typical light water reactor. Key considerations of this analysis are the stratification of the steam and helium, condensation deposition, and the flow patterns within the vessel.


Lane J.W.,Numerical Applications Inc. | George T.L.,Numerical Applications Inc.
International Topical Meeting on Nuclear Reactor Thermal Hydraulics 2015, NURETH 2015 | Year: 2015

Like many finite volume-based thermal-hydraulic codes used within the nuclear industry, GOTHIC™ uses a staggered grid formulation, where the control volumes for the momentum equations are offset from the control volumes for the mass and energy equations. This scheme is advantageous because it directly couples the pressure and velocity in a numerically stable solution, but it also creates challenges for formulating the convection terms. In particular, the momentum convected by an orthogonal component of the velocity is the most difficult to handle numerically and the formulation can influence results and stability. The effect of the finite volume formulation for the orthogonal convection terms in momentum equations is most apparent in a situation where the velocity vector is diagonal to the grid lines and there is a strong spatial gradient in the convected momentum. In a multiphase code like GOTHIC, these large gradients are typically due to large gradients in the phase volume fractions. Depending on the treatment of the orthogonal convection terms, these conditions can lead to a numerical momentum source in the phase that is being depleted as it moves across the grid. The momentum source was evident in a case with liquid injection (either continuous or droplets) into a gas filled volume at an angle to the grid. A particular example is drop injection from a spray nozzle. Although the observed momentum source is non-physical, this behavior had minimal impact on the overall results because the volume (and mass) of the accelerating fluid is diminishing with distance from the injection location. A modified approach for calculating the momentum convected by orthogonal velocity components has been developed and implemented in GOTHIC that offers an improved treatment for this unique scenario. Both first- And second-order accurate schemes are considered. This paper outlines the modified donoring schemes with an emphasis on the attributes that address consistency, stability, and phase appearance. Results from initial testing comparing the original and modified schemes for a simple and more detailed simulation are presented. GOTHIC™ incorporates technology developed for the electric power industry under the sponsorship of EPRI, the Electric Power Research Institute. © Copyright (2015) by American Nuclcar Society All rights reserved.

Loading Numerical Applications Inc. collaborators
Loading Numerical Applications Inc. collaborators