Development and Testing of TRACE/PARCS ECI Capability for Modelling CANDU Reactors with Reactor Regulating System Response

Abstract

The use of the USNRC codes TRACE and PARCS has been considered for the coupled safety analysis of CANDU reactors. A key element of CANDU simulations is the interactions between thermal-hydraulic and physic phenomena with the CANDU reactor regulating system (RRS). To date, no or limited development has taken place in TRACE-PARCS in this area. In this work, the system thermal-hydraulic code TRACE_Mac1.0 is natively coupled with the core physic code PARCS_Mac1.0, and RRS control is implemented via the exterior communications interface (ECI) in TRACE. ECI is used for coupling the external codes to TRACE, including additional physical models and control system models. In this work, a Python interface to the TRACE ECI library is developed, along with an RRS model written in Python. This coupling was tested using a CANDU-6 IAEA code coupling benchmark and a 900 MW CANDU model for various transients. For the CANDU-6 benchmark, the transients did not include RRS response, however, the TRACE_Mac1.0/PARCS_Mac1.0 coupling and ECI script functionality was compared to the previous benchmark simulations, which utilized external coupling. For the 900 MW CANDU simulations, all aspects of the ECI module and RRS were included. The results from the CANDU-6 benchmark when using the built-in coupling are comparable to those previously achieved using external coupling between the two codes with coupled simulations taking 2x to 3x less execution time. The 900 MW CANDU simulations successfully demonstrate the RRS functionality for the loss of flow events, and the coupled solutions demonstrate adequate performance for figure-of-eight flow instability modeling.

Modelling and Validation of CANDU Shim Operation Using Coupled TRACE/PARCS with Regulating System Response

“…Te RRS model was developed in the Python programming language and uses a Python ECI library built around the original Fortran-based ECI library. Tis coupled model has been previously evaluated against shorter transient events such as the fgure of eight fow oscillations (using a modifed system model without header interconnects) and a loss of fow transient initiated by a loss of Class IV power [12]. Tis section covers the additional work performed to model long-duration transients such as adjuster shim operation.…”

“…Te station design being simulated is a 900 MW class CANDU reactor, based on those used in previous coupled simulations of TRACE_Mac1.0 and PARCS_Mac1.0 [12]. When comparing preliminary results to station data, it was found necessary to update the core physics model for better consistency with the specifc CANDU 900 under analysis.…”

An Efficient and Robust ILU(k) Preconditioner for Steady-State Neutron Diffusion Problem Based on MOOSE

“…After the discretization of these nonlinear PDEs, it usually leads to a large-scale nonlinear algebraic equation problem [3]. Compared with the traditional methods, such as the operator splitting method and Picard iteration method [4][5][6], the Jacobian-free Newton-Krylov (JFNK) algorithm [7] is a powerful solver for the nonlinear equations due to its high-order convergence rate. The JFNK algorithm has been widely used in the newly developed nuclear reactor simulator [8][9][10][11], such as the MOOSE platform [12] and LIME platform [13,14].…”

An Efficient and Robust ILU(k) Preconditioner for Steady-State Neutron Diffusion Problem Based on MOOSE

“…The iterative method is a preferred option for the large-scale problem, since its smaller storage and computational cost compared with the direct matrix inversion method. Compared with the widely-used traditional coupling methods, such as operator splitting method and Picard iteration method [4][5][6] , Jacobian-free Newton-Krylov (JFNK) algorithm [7] is well-known due to its low storage and high-order convergence rate. JFNK algorithm has been widely used in the latest nuclear engineering simulation codes [8][9][10][11] .…”