IMPLEMENTATION OF TWO-GROUP INTERFACIAL AREA TRANSPORT IN A ONE-DIMENSIONAL COMPUTATIONAL ENVIRONMENT

摘要

A simplified one-dimensional computer code has been written which uses the two-fluid model to predict two-phase flows in large diameter vertical channels. Simplifying assumptions have been used to eliminate the energy equation and simplify the momentum equations. This code, written using MATLAB, uses a two-group momentum approach with two-group interfacial area transport to dynamically predict the development of the flow. This provides the ability to evaluate the interfacial area transport equation within a similar computational environment to that used in best estimate thermal-hydraulics analysis codes used in the nuclear industry and allows the evaluation of the performance of the transport equation, constitutive relations, bubble coalescence and breakup models, and other models using the same framework in which they will be used. The current version is limited to systems without phase change, i.e. where boiling and condensation are not significant. While the force-balance approach to interfacial drag has been used in the analyses presented, the code also includes a two-group drift-flux correlation and the implementation of the drag-coefficient approach recommended by Ishii and Hibiki as options within the code, allowing the comparison of various interfacial drag approaches. The code has been used to simulate the experiments performed by Shen et al in pipes with diameter of 0.2 m and length of 26.0 m. The code was able to duplicate the characteristic axial void fraction profile caused by the transition between the bubbly and cap-bubbly flow regimes in the experimental facility, showing that the two-group approach can be applied within the framework of existing best estimate codes. Further improvements to the accuracy of the code are expected if the bubble coalescence and breakup models are benchmarked based on a wider database and using the new approach to compute the flow development.