A new data processing tool for generating homogenized cross sections for the HEXBU-3D code
29th Symposium of AER on VVER Reactor Physics and Reactor Safety (2019, Energoland, Mochovce NPP, Slovakia)
fuel management issues
At Loviisa NPP, nodal code HEXBU-3D/MOD5 is utilized in designing the loading patterns as well as in the online monitoring of the reactor core performance. The HEXBU-3D code is based on two-group diffusion formalism and uses parametrized group constants as an input data. In the MOD5 version of the code, the feedback effects are modelled using second-order polynomials.
The current practice to create a group constant library for HEXBU-3D is to apply a miscellaneous set of various programs and scripts that carry out lattice calculation using CASMO-4E and, thereafter, post-process the obtained results into the required format. Improvement in this procedure was seen necessary and, therefore, a small-scale internal R&D program was started in 2019. The target of this project is to develop a completely new tool, XS-TOOL, for generation of group constant libraries for HEXBU-3D.
In this paper, the current status of the XS-TOOL development is documented. The code, written in Python language, is structured into modules including a general-purpose module for handling CASMO’s .cax output file. During the development work, special attention is paid on QA issues, e.g. source code style conventions and version handling, and traceability of the results created using the tool. The XS-TOOL software also includes features that enable the tool to be used for various data analysis purposes according to individual needs.
The paper documents the comparison results confirming that the XS-TOOL produces essentially identical results with the current production tools. The paper also demonstrates results of a spin-off study where limitations of the second-order feedback model were examined. Future plan is to expand the capability of the XS-TOOL for generation of MOD6 format library. In the MOD6 version of the HEXBU-3D code, the feedback effects are modelled using high-order polynomials making the MOD6 version suitable for dynamic analyses.