Nuclear Engineering and Design 243 (2012) 76–86 Contents lists available at SciVerse ScienceDirect Nuclear Engineering and Design jo u r n al hom epage : www.elsevier.com/locate/nucengdes Performance evaluation of a 3-D kinetic model for CANDU reactors in a closed-loop environment Lingzhi Xia a , Jin Jiang a,∗ , Hooman Javidnia a , John C. Luxat b a Department of Electrical & Computer Engineering, The University of Western Ontario, London, Ontario N6A 5B9, Canada b Department of Engineering Physics, McMaster University, Hamilton, Ontario L8S 4L7, Canada a r t i c l e i n f o Article history: Received 9 August 2011 Received in revised form 26 November 2011 Accepted 29 November 2011 a b s t r a c t The performance of a newly developed three dimensional (3-D) reactor kinetic model for CANDU reactor is evaluated in this paper. The evaluation is carried out in a closed-loop environment with the help of an existing reactor regulating system (RRS). The 3-D model is obtained based on a modal synthesis method, which represents the neutron flux distribution inside the reactor in both time and space. Throughout this paper, a special attention has been paid to compare the performance of the developed 3-D model with that of traditional coupled point kinetic models. The dynamic behavior of the reactor model in a practical load-following mode has also been examined. The accuracy of the model has been validated against actual plant measurements under both transient and steady-state conditions. Through the analysis and the simulation studies, it has convincingly demonstrated that the developed 3-D model has significant advantages over traditional coupled point kinetic models in terms of the improved accuracy and the higher resolution in modeling the reactor internal dynamics. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Computer programs can be found in every aspect of nuclear power plant from design to operation. For design and safety analysis of CANDU reactors, the commonly used codes for thermal- hydraulic, reactor physics, and LOCA analysis are: Canadian Algorithm for THErmal-hydraulic Network Analysis (CATHENA) (Hanna, 1998), Reactor Fuelling Simulation Program (RFSP) (Rouben, 1995), and Element Loss-Of-Coolant Accident (ELOCA) (Tayal et al., 1987), respectively. During plant operations, real-time computer algorithms have also been developed for on-line moni- toring and real-time regulation of the key system variables, such as neutron flux or reactor power. A good example is reactor regulat- ing system (RRS), which regulates the reactor power by means of adjusting the reactivity devices. Despite of the above design, analysis and operational tools, one area that seems to be left out, or at least with less attention paid, is software design tools for CANDU reactor control system design and analysis. For many years, point kinetic models have been used to approximate the dynamic behaviors of the reactor for control system design and analysis. The main advantage of a ∗ Corresponding author. Tel.: +1 5196612111x88320; fax: +1 5198502436. E-mail addresses: lxia4@uwo.ca (L. Xia), jjiang@eng.uwo.ca (J. Jiang), hooman@javidnia.org (H. Javidnia), luxatj@mcmaster.ca (J.C. Luxat). point kinetic model is its ability to represent the essential dynamic relationships of neuron flux in a simple manner, which makes the analytical design of control system straightforward. Unfortunately, the price to pay for this simplicity is low resolution in reactor modeling. Hence, one cannot differentiate flux distributions at dif- ferent locations in the reactor. A more detailed reactor dynamic model is developed by using 14-coupled point-kinetic equations (Tiwari, 1996). Subsequently, a design tool has been developed in MATLAB/SIMULINK environment to provide a user-friendly control system design environment (Javidnia et al., 2009). Even though the 14-coupled point-kinetic equation provides much detailed representation of the reactor core, it is still assumed that the neutronic properties within one zone in the core are uni- form. The neutron dynamic behavior in the zone corresponds to a fixed flux distribution shape. In other words, the spatial information of the core is approximated by only 14 distinctive zones. To over- come this inherent limitation associated with point kinetic model approaches, a new reactor model by modal synthesis method has been developed recently (Xia and Jiang, 2011) for CANDU reactors. This technique treats the neutronic kinetic distribution in a core as a weighted sum of responses of a finite number of three-dimensional (3-D) flux modes. The shapes of these flux modes can be deter- mined through steady-state analysis of the reactor core. Therefore, they are functions of the core location, but are completely indepen- dent of time. On the other hand, the weighting functions are the sole functions of time. Consequently, the neutronic distribution at 0029-5493/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.nucengdes.2011.11.034