Spure Dike

Experiment Setup

The laboratory flume experiment of Rajaratnam and Nwachukwu (1983) consists of a rectangular 37 x 0.92 m flume with open-channel flow and a thin plate of length 0.1524 m used to simulate a groin-like structure. Here the numerical model is compared to experiment A1 of Rajaratnam and Nwachukwu (1983). For this case, the flow depth and discharge were 0.189 m and 0.0453 m³/sec, respectively.

Model Setup

The computational grid size is 152 x 36 and has a variable grid resolution of 0.01 to 0.05 m (Figure 1). The mesh was refined manually near the structure, near the walls within the recirculation zone behind the structure. A constant flux was specified at the inflow boundary and a zero water level at the outflow boundary.

Figure 1. Computational grid for experiment A1 of Rajaratnam and Nwachukwu (1983) .

Results

Flow velocities were measured down stream of the flow structure along 4 transects in the transverse direction and at two arbitrary levels or heights y/y0=0.03 and y/y0=0.85. The velocities from y/y0=0.85 are compared to the calculated depth-averaged velocities and are shown in Figure 2.

Figure 2. Comparison of measured and computed transect

Figure 3.Computed water levels and currents

Table 1. Goodness of fit statistics for U-velocity, m/s

	Transect 1	Transect 2	Transect 3	Transect 4
RMSE	0.0504	0.0690	0.0557	0.0627
RMAE	0.0239	0.0725	0.0884	0.1038
R2	0.978	0.951	0.975	0.993

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

References

• Rajaratnam, N. and Nwachukwu, B.A. (1983). "Flow near groin-like structures." Journal of Hydraulic Engineering, 109(3), 463-480.

---

---

Test Cases

Documentation Portal