Abstract:
The applicability of experimental and numerical models used for the solution of dam-break flows is vital for better dam projects and also in preventing related accidents. The high cost and the time-consuming nature of laboratory studies require consistency in the investigation of numerical models. In this study, the propagation of a flow using a fluid with a different density from that of normal water in the reservoir was investigated both experimentally and numerically. Salt water was preferred as a Newtonian fluid in order to observe the propagation of flows in different density after a sudden break. A small-scale channel was constructed and laboratory data were obtained using image processing techniques. For the numerical model, Smoothed-Particle Hydrodynamics (SPH) method and Reynolds Averaged Navier-Stokes (RANS) equations solved by Flow-3D software, were applied. Flow depth changes were observed in the reservoir and the downstream. The data obtained from all methods were compared with each other. The results of two numerical simulations point out that the disagreements on graphs in the time evolutions of the fluid levels in the SPH increase due to turbulence effects, whilst, these differences decrease in the RANS equations solved by Flow-3D software. Consequently, since the SPH provides taking the measures and developing intervention strategies to reduce the risks connected to the evolution of dam-break flows, it is thought that future validation studies of the model will be require with the use of data observed in this field.