Advanced Multi-Phase Flow Laboratory

                       Department of Mechanical and Nuclear Engineering


Current Projects

TRACE with Dynamic Model for Interfacial Area


Research Supported by the United States Nuclear Regulatory Commission




    In solving two-phase flow problems, the interfacial area concentration, ai, is one of the most important parameters to estimate accurately. This surface area is directly related to the amount of mass, momentum and energy that can be transferred between two interacting phases. Currently, nuclear reactor system analysis codes, such as TRACE (NRC) or RELAP5-3D (DOE), rely on flow regime based algebraic closure relations. These relations do not reflect the dynamic evolution of the flow and can produce numerical issues at the flow regime boundaries. To address these issues, the current work continues the previous evaluation of the one-group interfacial area transport equation (IATE) in TRACE, TRACE-T1. Moreover, the current work seeks to implement an adiabatic two-group IATE, TRACE-T2, that accounts for both small bubbles and large bubbles.



Figure 1. Types of bubble interations.

    To date, 50 different experimental air-water, two-phase flow conditions in various pipe sizes in both vertical-upward and vertical-downward configurations have been assessed with TRACE-T1. Examples of the assessments are shown in Figure 2. Here, the results are also compared against the results obtained by the conventional flow regime based relations (TRACE-NT). Since TRACE-NT employs a pressure based approach, for downward flow the interfacial area will always decrease as pressure increases along the flow direction, shown on the left. However, for certain turbulent flow conditions, the interfacial area actually increases due to bubble breakup, as captured by TRACE-T1, but not TRACE-NT. In upward flow, TRACE-NT always predicts an increase in ai due to the pressure drop along the flow direction. However, as can be seen on the right, TRACE-T1 can capture a decrease in ai due to the promotion of coalescence mechanisms. Clearly, the IATE can greatly enhance the accuracy of the code predictions by accounting for additional physics of the flow.



Figure 2. Characteristic predictions of ai by TRACE-T (with IATE) and TRACE-NT (flow regime based); error bars are 20%. Left: 2.54 cm co-current downward, jg= 0.404 m/s and jf= 3.110 m/s. Right: 20.32 cm upward, jg= 0.300 m/s and jf= 0.420 m/s.