Sudden Expansion

UNDER CONSTRUCTION

Experimental Setup

The flume experiment consisted of a rectangular flume 18 m long with an inflow section 0.6 m wide which expands suddenly into a wider section 1.2 m wide. The experiment parameters are shown in Table 1.

Table 1. General Settings for Sudden Expansion Case

Parameter	Value
Discharge	0.0385 m^3/s
Downstream water level	0.0 m
Manning's coefficient	0.015

Model Setup

Figure 1. Computational grid.

The streched 3-level quadtree CMS-Flow grid is shown in Figure 1. The grid has a resolution between 0.03 and 0.45 m. A flux boundary condition is specified at the inflow boundary and a constant water level boundary condition is applied to the downstream boundary. The model setup parameters are shown in Table 2.

Table 2. Model

Parameter	Value
Numerical scheme	Implicit
Time step	30 s
Simulation Duration	1 hr
Ramp time	0.5 hr

Results

Figure 2. Comparison of analytical and calculated water surface elevations and bed elevations.

A comparison between the analytical solution for water levels is compared to the calculated water levels in Figure 2. As shown in the goodness of fit statistics in Table 3, the model results agree well with the analytical solution. The minimum water level is captured well, however there is a small shift in the location of the water level drop over the bump toward the downstream direction.

Table 3. Goodness of fit statistics for the current velocities

Statistic	x = 0 m	x = 1 m	x = 2 m	x = 3 m	x = 4 m	x = 5 m
RMSE, m/s	0.0105	0.0260	0.0294	0.0386	0.0775	0.0584
NMAE, %	1.28	0.72	0.56	3.19	4.33	0.45
R^2	0.789	0.995	0.990	0.989	0.936	0.980
Bias, m/s	-0.0083	-0.0046	0.0035	0.0187	-0.0017	0.0022

• For a definition of the goodness of fit statistics see Goodness of fit statistics.

References

• Xie, B.L. (1996). "Experiment on flow in a sudden-expanded channel," Technical Report, Wuhan University of Hydraulic and Electric Engineering, China.

---

Test Cases

Documentation Portal