Abstract
Partial convergence of CMFD can help to stabilize multiphysics iteration schemes. In this paper, an efficient multiphysics iteration scheme with near-optimal partially convergent CMFD implemented in MPACT is presented. In the new scheme, the feedback intensity of the problem is automatically estimated, and the relative convergence of CMFD solver is adjusted accordingly. Numerical results show that MPACT with near-optimal partially convergent CMFD can have almost the same convergence rate in problems with feedback as those without feedback. For the problems tested here the run time may be reduced by more than 20% and up to 49% compared with that of current MPACT.
Highlights
The Coarse Mesh Finite Difference (CMFD) [1] method is a standard Nonlinear Diffusion Acceleration (NDA) method for accelerating numerical transport calculations in reactor physics [2,3,4]
Our previous research [8] has shown that partially convergent CMFD can stabilize the multiphysics iteration scheme using a relaxation-free method that determines the near-optimal partial convergence of CMFD according to the feedback intensity of the problem
The near-optimal partially convergent methods (MGPC/MSEDPC) adjusts the convergence of the CMFD solver according to the problem dependent feedback intensity, and allows MPACT to have almost the same number of outer iterations to converge the problem at any power level without the relaxation technique
Summary
The Coarse Mesh Finite Difference (CMFD) [1] method is a standard Nonlinear Diffusion Acceleration (NDA) method for accelerating numerical transport calculations in reactor physics [2,3,4]. Our previous research [8] has shown that partially convergent CMFD can stabilize the multiphysics iteration scheme using a relaxation-free method that determines the near-optimal partial convergence of CMFD according to the feedback intensity of the problem. This method was shown to have a theoretical spectral radius comparable with NDA in problems without feedback, and nearly the same behavior for problems with feedback. The implementation details are described and numerical results comparing the performance with the standard relaxed Picard iteration presently used in MPACT for a wide range of multiphysics problems are presented
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have