Transient compressible simulation of a high-speed radial compressor using the CATHARE system code

Authors

Valentin Cottarel  - French Alternative Energies and Atomic Energy Commission (CEA), France
Laura Matteo - French Alternative Energies and Atomic Energy Commission (CEA), France
Gédéon Mauger - French Alternative Energies and Atomic Energy Commission (CEA), France

Abstract

The French Alternative Energies and Atomic Energy Commission (CEA) works on Brayton cycles modeling and design for the reactor energy conversion systems. Within this framework, the Code for Analysis of Thermal-Hydraulics during an Accident of Reactor and safety Evaluation (CATHARE) is used as computational tool is used to perform thermal-hydraulic calculations. It is a transient two-phase flow six-equation model closed by physical closure laws and solved using a Newton-Raphson iterative method. It is able to reproduce various fluids (such as water, sodium, hydrogen or air) with their real properties like their compressibility and to simulate different flow regimes. Diverse possibilities for simulating turbomachinery exist at several scales in the code: a nodal non-predictive 0D model and a 1D transient predictive model. This 1D model is fast in terms of CPU time, is used for pre-design studies and can be included in a full system scale computation without deteriorating the global calculation time. A complete description of the modeling can be found in the Ph.D thesis of L. Matteo. The development of the model includes various validation cases. It is able to predict the performances of a machine on a wide flow rate range (typically [10%QN - 150%QN]), at various rotational speeds with a relative error generally lower than 10%. This 1D model considers a mean streamline adopted by the fluid on each element of the machine. As a result, it allows to have an internal representation of the machine as all the thermal-hydraulic quantities can be post-processed on each mesh using the CATHARE code (pressure and velocity profiles can be obtained on each element of the machine, fixed and rotating parts). This model has so far been mostly applied to hydraulic machines and to a one low-speed compressor where compressibility effects are not clearly visible. The aim of this work is to extend the model on a high-speed compressor to match the needs of the Brayton cycles. For example, it could be used to model the low pressure compressor of the Pebble Bed Micro Model (PBMM) where the rotational speed of the machine is around 60000 rpm. The studied turbomachine is the NASA's CC3 compressor. This is a 4:1 pressure ratio centrifugal compressor with a nominal rotational speed of 21789 rpm. The geometry is fully described by McKain and Skoch as well as the experimental performances. The first task is to collect all the data necessary for the model (geometric specifications, hydraulic parameters, experimental conditions...) and to establish for verification purposes a calculation under ideal conditions to compare it to Euler's theory. Then a work is carried out to predict the performances in real conditions, on the various rotational speeds available in order to check the similarity laws. The developments and improvements useful for the modeling of this particular machine are described, especially the the management of a compressible fluid.