Elsevier

Annals of Nuclear Energy

Volume 159, 1 September 2021, 108290
Annals of Nuclear Energy

Accuracy and efficient solution of helical coiled once-through steam generator model using JFNK method

https://doi.org/10.1016/j.anucene.2021.108290Get rights and content

Highlights

  • The advanced JFNK method is used to solve the highly nonlinear H-OTSG system.

  • An efficient preconditioner is used, which is derived from semi-implicit method.

  • Technologies of phase generation/disappearance and the global strategy are used.

  • A new H-OTSG code is developed and applied to analyze H-OTSG in HTR-10.

Abstract

The Helical coiled tube Once-Through Steam Generator (H-OTSG) is key equipment in the High-Temperature Gas-cooled Reactor (HTGR) system, where the fission energy is transferred from the helium side to the water side, together with the complicated heat transfer and phase change phenomenon. In detail, the water is heated to the water-steam two-phase flow by the hot helium and further to the superheated steam. As a result, the H-OTSG is a multi-phases, multi-physics, and multi-domain highly nonlinear system due to the complex constitutive equations in the two-phase flow model. The accuracy and efficient solution of this highly nonlinear system is a key point for the successful simulation of the H-OTSG. In this work, the advanced Jacobian-Free Newton Krylov (JFNK) method is used instead of the traditional semi-implicit iteration method to pursue high computational performance. An efficient physical-based preconditioner is proposed, which is derived from the original semi-implicit based method. Moreover, the numerical technologies of the phase generation/disappearance and the global convergence algorithms of JFNK method are implemented and discussed. The two-phase drift-flux model is used to describe the two-phase flow in the waterside and the second-order accuracy temporal and spatial numerical scheme is employed to discrete the conservation equations. A new H-OTSG simulation code based on JFNK method is developed in this work and applied to analyze the steady and transient state behavior of the H-OTSG in the 10 MW high-temperature gas-cooled reactor (HTR-10). The simulation results show that the accuracy of the newly developed H-OTSG simulation code agrees well with the design data under different operating conditions. Furthermore, the computational performance of the JFNK method with the physical-based preconditioner is the best compared with the traditional Picard iteration method, Newton–Krylov method and JFNK method without preconditioner.

Introduction

Pebble-bed High-Temperature Gas-cold Reactor (HTGR) is one of the 4th generation nuclear power plants, featured with a high temperature at the reactor outlet and high thermal efficiency. As a key equipment in HTGR system, the Helical coiled tube Once-Through Steam Generator(H-OTSG) couples the primary circuit and the secondary circuit, transferring the fission energy from the hot helium at the first side to the water at the second side. Compared with the traditional U-tube natural circulation steam generator, the water in the H-OTSG is not only heated reaching the water-steam two-phase flow but also further reaching the superheated steam to pursue higher thermal efficiency. The centrifugal forces acting on the fluid inside the helical ducts tend to laminarize the flow, enhance the heat transfer coefficient (Xia et al., 2017). The accurate simulation of the H-OTSG behavior, especially the complicated the two-phase flow phenomenon, is an important step for a successful HTGR design, as well as the safety analysis. A continuous effort has been carried out to achieve the accurate simulation of the steam generators in the HTGR plant.

The fully three-dimensional CFD-based simulation is a useful numerical tool to provide detailed information in the steam generator (Ferng and Chang, 2008), but still suffering from the too-exhausted computational cost. The one-dimensional model is the most widely used numerical tool for steam generator simulation in the engineering application due to its practicality and efficacy. Numerous studies have been carried out on inveatigating single-phase and two-phase pressure drop and heat transfer characteristics of H-OTSG. BLAST is a specially designed one-dimensional computational code for the H-OTSG in HTGR plant (Hedrick and Cleveland, 1976), which has been successfully coupled with the reactor core simulation codes, such as THERMIX (Zheng et al., 2012) and TINTE (Gerwin et al., 2009), and has been widely used in HTGR engineering design and safety analysis. In the BLAST code, the homogeneous flow model is used to describe the two-phase flow behavior in the waterside, which assumes that the steam and water share with the same velocity and the velocity difference between the two phases could not be distinguished. Moreover, the conservation equations of the helium side, tube component and waterside are sequentially solved by the traditional semi-implicit coupling method in the BLAST code. Therefore, the coupling information from the waterside is lagged result in inconsistent convergence in each time step, which leading to the material Courant time step limit. To assess the design decisions and investigate the transient behavior, a lot of work on H-OTSG has been conducted. Ahn et al. (2017) validated the H-OTSG model using the system transient analysis program TASS/SMR developed by Lee et al. (2009). Xiaowei Li developed a lumped-parameter H-OTSG model based on a movable boundary method to perform the steady-state calculation and the simulation results agree well with design data (Li et al., 2011). Zhe Dong proposed the simplified one-dimensional H-OTSG models as a component in the whole HTGR system for designing the automatic control system (Dong et al., 2020). Reactor safety analysis codes based on the two-phase flow model may also have been capable to simulate the behaviors of H-OTSG after moderate modification, such as RELAP5 (U.S. NRC, 1995a), TRAC (U.S. NRC, 2010), RETRAN (Moore et al., 1977) and TRACE (U.S. NRC, 2010). However, these safety analysis codes use the traditional semi-implicit coupling method and the time step is also limited by the material Courant number.

To the authors’ best knowledge, most existing work focus on the detailed analysis of the two-phase flow behavior within H-OTSG. However, only a few attempts are made from the viewpoint of the numerical solution method, as well as their efficiency and accuracy. The fully implicit, super-linear convergent Jacobian-Free Newton Krylov (JFNK) method is used in this work to pursue higher computational performance. JFNK method is an advanced fully implicit coupling method and has been successfully applied in the multiphase flow (Zou et al., 2016) and multi-physics coupling issues (Zhang et al., 2018). Different from the traditional semi-implicit method, the highly nonlinear conservation equations of the helium side, tube component and waterside are solved in a tightly nonlinear form as a whole system in JFNK method. All the physical quantities are updated simultaneously and consistent convergence in time. In this work, the two-phase drift-flux model is used to consider the velocity difference between the two phases in the waterside. The second-order accuracy temporal and spatial numerical scheme is utilized to achieve a higher numerical accuracy, compared with the first-order accuracy scheme in most existing codes. Furthermore, please note that, JFNK method could provide a powerful mathematical freamework which has the potential to solve the whole HTGR coupled system within one uniform algorithm (Zhang et al., 2018). Several efforts have been made in our team to solve the neutron kinetic and thermal hydraulic coupled issue by JFNK method (Zhang et al., 2018, Zhang et al., 2019, Wu et al., 2020, Lu et al., 2018), and we plan to extend the coupling system from the reactor core to the whole primary circuit, and further to the whole HTGR nuclear power plant. In this work, we focus on the prediction of H-OTSG behavior using JFNK, which is an important and necessary part for the whole HTGR coupled system simulation under the uniform JFNK framework. The main contributions of this work are:

  • An efficient physical-based preconditioner is proposed for the H-OTSG model, which is derived from the original semi-implicit based method. The preconditioner plays an important role in the JFNK method to ensure high computational efficiency.

  • The numerical techniques of the phase generation/disappearance and the global convergence algorithms of JFNK method are implemented and discussed for the H-OTSG model.

The remainder of this paper is organized as follows: The conversation equations and structure relationships are described in Section 2; Fully implicit JFNK method and construction of precondition matrix are illustrated in Section 3; Numerical results and discussion of H-OTSG model are given in Section 4; Finally, conclusions and future work are provided in Section 5.

Section snippets

Descriptions and assumptions in the steam generator

The H-OTSG is a helical coil heat exchanger, where the hot helium flows on the heat transfer tube surface and the secondary side feed water flows inside the helically coiled tubes. After a countercurrent heat transfer, the single-phase water is heated to superheated steam by the hot helium. Moreover, the helium in the primary side, the metal helical coil tube, and the secondary side feed water are coupled together due to the conjugate heat transfer process.

Several widely used assumptions are

Fully implicit discretization of governing equations

In most existing thermal–hydraulic codes, the finite volume method(FVM) with a staggered grid is commonly applied for the spatial discretization. The scalar variables such as pressure, void fraction and internal energy are obtained at the mesh cell center, while the vector variables velocities are obtained at the mesh cell faces. The staggered grid mesh is selected because it is flexible to handle the two-phase equations in primitive forms and it is compatible with most existing system analysis

Result

A thermal–hydraulic code for once-through steam generator has been developed using the fully implicit JFNK method. The steady-state and transient behaviors for the H-OTSG in HTR-10 reactor have been simulated to analyze the performance of the code. The key parameters and initial/boundary conditions for this steam generator are summarized in reference Ju et al. (2001). An efficient numeric library-Portable, Extensible Toolkit for Scientic Computation (PETSc) is implemented in this work (Balay et

Conclusion

The H-OTSG is the largest heat exchanger in the HTGR system, together with the complicated conjugate phase change heat transfer phenomenon. In this work, the high-order time and spatial integration scheme with a staggered grid are used in the discretization of HTR-10 H-OTSG. The JFNK method is applied to solve fully implicit nonlinear equations. Deriving from the semi-implicit discretized governing equations, a conceptually simple and computational efficient preconditioner is constructed. The

Data availability

The data used to support the findings of this study are available from the corresponding author upon request.

CRediT authorship contribution statement

Yingjie Wu: Conceptualization, Methodology, Software, Writing - original draft. Baokun Liu: Methodology, Software. Han Zhang: Conceptualization, Methodology, Writing - review & editing. Kaijie Zhu: Software. Boran Kong: Software. Jiong Guo: Writing - review & editing. Fu Li: Conceptualization, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work is supported by Beijing Natural Science Foundation 1212012, Chinese National Natural Science Foundation Project 11505102 and 11375099, Chinese National S&T Major Project 2018ZX06 902013, and IAEA CRP I31020.

References (49)

  • L. Zou et al.

    Applications of high-resolution spatial discretization scheme and jacobian-free newton-krylov method in two-phase flow problems

    Ann. Nucl. Energy

    (2015)
  • C. Frepoli et al.

    Notes on the implementation of a fully-implicit numerical scheme for a two-phase three-field flow model

    Nucl. Eng. Des.

    (2003)
  • D.A. Knoll et al.

    Jacobian-free newton-krylov methods:a survey of approaches and applications

    J. Comput. Phys.

    (2004)
  • Z. Wu et al.

    The design features of the htr-10

    Nucl. Eng. Design

    (2002)
  • B. Liu et al.

    Finite difference jacobian based newton-krylov coupling method for solving multi-physics nonlinear system of nuclear reactor

    Ann. Nucl. Energy

    (2020)
  • Hedrick, R.A., Cleveland, J.C., 1976. Blast: a digital computer program for the dynamic simulation of the high...
  • Gerwin, H., Scherer, W., Lauer, A., 2009. Tinte-a two-dimensional code for reactor dynamics. Technical report, Berichte...
  • Ahn, S.J., Yang, S.H., Chung, Y., Bae, K.H., Lee, W.J., 2017. Gap conductance model validation in the tass/smr-s code....
  • X. Li et al.

    Transient pressure analysis for the reactor core and containment of a htgr after a primary loop pressure boundary break accident

    Chin. Sci. Bull.

    (2011)
  • Z. Dong et al.

    Cascaded htgr power-level control only by regulating primary helium flow rate

    IEEE Trans. Nucl. Sci.

    (2020)
  • U.S. NRC. 2010. Trace v5.0 theory manual-field equations, solution methods, and physical models. Technical Report, U.S....
  • Moore, K.V., Hughes, E.D., McClure, J.A., 1977. Retran: a program for one-dimensional transient thermal-hydraulic...
  • Zou, L., Zhao, H., Zhang, H., 2016. Numerical implementation, verification and validation of two-phase flow...
  • H. Zhang et al.

    An assessment of coupling algorithms in htr simulator tinte

    Nucl. Sci. Eng. J. Am. Nucl. Soc.

    (2018)
  • Cited by (21)

    • A three-region movable-boundary helical coil once-through steam generator model for dynamic simulation and controller design

      2023, Nuclear Engineering and Technology
      Citation Excerpt :

      Arda and Holbert [16] adopted an OTSG for simulation of a small modular reactor, which is limited to have identical feedwater and steam flow. Recently, more accurate models have been developed [17–20], which are good for thermal-hydraulic characteristics analysis. There are also many studies on OTSG simulations using CFD methods [21], but these models cannot be implemented for controller design, and thus will not be discussed here.

    • A modified JFNK with line search method for solving k-eigenvalue neutronics problems with thermal-hydraulics feedback

      2023, Nuclear Engineering and Technology
      Citation Excerpt :

      A high-performance preconditioner is crucial for Krylov iteration method to solve the linear system efficiently in Step.2.1, as shown in Algorithm 1. JFNK method as well as its variants have been widely used in complicated thermal-hydraulics successfully [18–22], computational fluid dynamics [23–26] and multi-physics coupling issue [10–13,27–30] in nuclear engineering and other scientific communities. More detailed information about the JFNK methods could be found in the reference [31].

    • An improved-MPGA and its application in OTSG load reduction characteristics optimization

      2023, Annals of Nuclear Energy
      Citation Excerpt :

      Since the fluid phase transition exists on the secondary side heat exchanger tube of the Once-through Steam Generator (OTSG), its dynamic characteristics are extremely susceptible to changes in operating parameters. Although the study of the thermal hydraulic behavior of the secondary side of the OTSG is becoming more and more complete and accurate (Chen et al., 2019; Ding et al., 2020; Wu et al., 2021; Xia et al., 2015; Zhang et al., 2020), it is a challenge for the nuclear power facilities in service to exploit its potential to improve the operating characteristics. Optimization methods may provide guidance and help.

    • A multi-level nonlinear elimination-based JFNK method for multi-scale multi-physics coupling problem in pebble-bed HTR

      2022, Annals of Nuclear Energy
      Citation Excerpt :

      It is challenging to solve such a complicated multi-physics, multi-scale coupled system. The recent development of Jacobian-free Newton Krylov (JFNK) algorithm attracts extensive attention and has been applied in reactor engineering community (Wu et al., 2021; Esmaili et al., 2020; Zhang et al., 2019, 2022; Wu et al., 2020; Niu et al., 2021; Yang et al., 2020) and other communities successfully (Taverniers and Tartakovsky, 2017; Liao et al., 2021), due to the fully-implicit scheme, super-linear convergence rate and stable convergence capability. However, most existing works focus on the traditional multi-physics coupling issue (Liu et al., 2020; He et al., 2020; Park et al., 2010; Balestra et al., 2021), where all the physical fields share the same computational domain.

    View all citing articles on Scopus
    View full text