مجله هیدرولیک
سال دهم شماره 3 (پاییز 1394)

Formation of bed forms, such as dunes, is a prominent feature in natural open channel flows with sediment transport. Knowledge of turbulent flow over sand dunes is needed to understand the interaction between the flow, the bed forms and the associated sediment transport. This study presents results of an experimental work aimed at investigating the effect of different dunes and vegetated banks on flow velocity, Reynolds stresses and turbulence intensities. Experimental measurements were made over fifth and sixth dunes in a series of ten of 2-dimensional asymmetric gravel dunes which had dune height of 4 and 8 cm, respectively. These dunes had a mean wavelength of 100 cm, mean lee slope angles of 28º, and width equal to the flume width. Rice stems with a median diameter of 2.7 mm and a stem distribution density of 400 stem/m were used to cover the flume banks for simulating vegetation on riverbanks. From experimental measurements, the effect of different heights of crest of gravel dunes on the characteristics of flow in vegetated flume; and the reattachment points for different heights of crest of gravel dunes were investigated.

Estimation of earthfill dam overtopping risk is an important issue in computing the dam height and spillway capacity. However, estimation of dam overtopping risk using Monte Carlo simulation and considering all effective random variable is very time consuming and is an ideal variant in assessment of dam overtopping risk. In this respect, due to the mentioned weakness of regular Monte Carlo simulation, the low number of random variables were used in previous studies. In this study, new approach for assessment of dam overtopping and reduction of simulation time was developed using rainfall threshold overtopping curve and Monte Carlo simulation. Result show that the relative error of estimated risk between regular and conditional (used in this study) Monte Carlo is 6 percent but the time of simulation was reduced about 29 hours. Hence, the proposed methodology (conditional Monte Carlo) respect to regular approach is more efficient in estimation of Jamishan dam overtopping risk.

The use of Pareto dominance for evaluating the solutions has been the mainstream in evolutionary multi objective optimization for the last two decades. An alternative is multi objective evolutionary algorithm based on decomposition (MOEA/D) which uses scalarizing the objective functions. In this paper, decomposition strategies are developed for water distribution network (WDN) design problems by integrating the concepts of genetic algorithm (GA) within the MOEA/D framework. The proposed algorithms are then compared with the two well-known non-dominance based MOEAs: NSGA2, and SPEA2. This comparison is made by plotting the Pareto fronts and evaluating the hyper-volume and two-set coverage metrics across two large scale WDN design problems. Experimental results show that MOEA/D outperform the Pareto dominance methods in terms of both non-domination and diversity criteria. It suggests that decomposition based multi objective evolutionary algorithms are very promising in dealing with real-world water engineering optimization problems.

Smoothed Particle Hydrodynamic method is a Lagrangian numerical particle method, best fitted for free surface hydrodynamics. SPHysics is one of the most well-known computer models based on SPH. However, this open source code is limited to one-phase Newtonian fluid problems. In this research, after developing the standard serial SPH code to a two-phase non-Newtonian model, the movable bed dam break problem has been modeled. An added pressure term is used in the pressure equation of state. Afterward, the model has been used for various problems and the validity of results was evaluated by related numerical models and experiments. Finally, the model has been challenged with modeling of dam break movable bed material by non-Newtonian Bingham model and compared with MPS method and experimental results.

A wide range of natural phenomena is modeled by Saint-Venant equations or the dynamic wave model, such as: dam breaks and flood routing. Flood routing is a mathematical method to forecast the changes of the volume, velocity and the shape of the flood wave as a function of time and space. Knowing flood intensity and its wave propagation manners plays a substantial role in minimizing human and financial disasters. A fairly significant change of dependent variable within a short distance is a primary criterion of those phenomena. Therefore, there are often non-physical and non-uniform oscillations in the result of their numerical model, which could entirely invalidate the solution which is obtained. Hence, in addition to an efficient discretization method, some measures must be utilized to reduce cited error and prevent unrealistic results. In this investigation the dynamic wave model was discretized by Finite Volume Method, which is a novel and absolutely reliable method. Then a straightforward algorithm was introduced to solve discrete equations set which prevents spurious numerical oscillations by putting ULTIMATE strategy in use. Furthermore, take advantages of different schemes is possible in innovative algorithm and it could be implemented in various coding programs. Finally, to examine discrete equations set and proposal algorithm, a code was written in MATLAB and its outputs were compared with Chow et al. (1988). Comparison demonstrated that discrete equations set were absolutely correct and the presented algorithm is efficient and well-established.

Hydraulic simulation of the "dam break" phenomenon is crucial in estimating the amount of losses, aid and rescue measures downstream of the dam. An example of such simulation is the prediction of the peak flood discharge caused by the break of concrete dams. In this study, the phenomenon of "dam break" is modeled. By considering the surrounding morphology, flood routing and flood zoning were also performed. In the present study, attempts have been made to compare the results obtained from the software “MIKE 11" with those from the analytical solution. Besides, the flood hydrographs caused by dam break and experimental equation were also compared. The results of the present study indicates that the impact of the peak flood discharge caused by the break of Dez Dam is likely to affect only 30 kilometers downstream of the dam and is effective only at the adjacent areas of the great Dez Dam around the city of Dezful. Furthermore, the results revealed that if Dez Dam breaks, floods of 60,000 and 11,000 cubic meters per second will reach Dezful and Ahvaz cities respectively.