simulation-optimization

Human is under threat due to the shortage of fresh water, its deteriorating quality, and rising water stress around the world. Proper management of water resources is very important to ensure the sustainable social and economic development in any country in the world. In order to sustain freshwater resources, they must be used efficiently without compromising the needs of future generations. The present study was conducted to ensure the sustainable management of groundwater resources. For this purpose, four indicators of vulnerability, resilience, reliability, and combined stability, each of which includes aspects of sustainability, were used as the objective function of management issues. Using the combined simulation-optimization model of Namdan plain aquifer and the ant colony optimization algorithm, a decision model was created. The results indicate the effectiveness of four indicators of vulnerability, resilience, reliability and combined stability to increase groundwater level and positive changes of water storage. So that the aquifer storage change under each scenario was 11.7, 13, 12.3 and 12.4 million cubic meters, respectively. According to the results obtained from four management issues, analyzed in this study, the most important effect in improving the aquifer with a significant reduction in pumping from well while the such reduction should be measured at the aquifer level.

In this research, for the first time, a simulation-optimization system dynamic-based model with Anylogic software for reservoir eutrophication management is presented. Thus, the relationships of the elements related to eutrophication were formulated in different layers of Karkheh Dam and in relation to each other in the system dynamics form. comparing the presented results with the results of the CE-QUAL-W2 and Vensim simulation models over a one-year period showed that the value of the correlation coefficient (𝑅2) for the variables phosphate, ammonium, nitrate, Dissolved oxygen and Chlorophyll-A between the two models of Anylogic and CE-QUAL-W2 is equal to 0.89, 0.81, 0.78, 0.75 and 0.86, respectively, and the value of this coefficient for the mentioned variables is equal between the two models of Anylogic and Vensim. With 0.98, 0.94, 0.90, 0.93 and 0.97, which shows the high correlation between the results of Anylogic model with the two mentioned models. Then, by combining the developed model in Anylogic environment with a simulated annealing optimization model, a simulation-optimization model was developed to determine the optimal values of water releases from different layers of reservoir in 15 years’ horizon considering the quantitative and qualitative objectives. Here, the RMSE between the results of the two models for the lower, middle and upper layers of the reservoir is 0.004, 0.007 and 0.010, respectively, and 𝑅2 is 0.91, 0.9 and 0.79. The results indicated that the results of the two models were similiar. This study showed the effectiveness of the proposed model in the management of dam eutrophication.

Uncertainty analysis is an inseparable step in the process of hydrological modeling. Quantitative assessment of the uncertainty in the simulation model outputs and its parameters, thereby increasing confidence in the results of modeling and understanding of the sources of uncertainty. Due to the increasing use of groundwater management model and predict the behavior of the aquifer, this research is seeking to analyze the uncertainty in quantitate-qualitative aquifer simulation and its effect on optimization results. Using SWAT hydrologic model, the amount of recharge is specified and entered into MODFLOW groundwater flow model and MT3DMS transmission model. In this research, the DREAM(zs) algorithm (based on Monte Carlo Markov chain algorithms) was used to examine the uncertainty of MODFLOW model parameters. Then by linking the model with MOPSO, the optimum head and salinity are obtained in the aquifer. The results show that the accuracy of the inputs of the model leads to the desirability of the results in relation to the intended purpose of reducing the water table.

In this study, the simulation of the BORKHAR plain aquifer located in north-east of Isfahan was done for the estimation of the hydrodynamic coefficients and for the preparation of the prediction and management model with the purpose of the study of the water table surface situation in the next years. The study of the geological situation of the plain and the report of the drilling of its exiting wells indicated that the BORKHAR plain has two kinds of aquifer: confined and unconfined. According to the field data related to the unconfined aquifer, a part of this aquifer was selected for the mathematical simulation. The calibration model for the estimation of the K and Sy. coefficients was done by dividing the plain into four geohydrologic units and by using the PEST, a module of the MODFLOW model. The situation of the water table level during 1380-1390 was studied according to the different management options by the calibration model. If the trend of the exploitation doesn’t change during the next ten years, we will confront with a maximum drop of 48 meters. As a practical way for preventing from this drop, it was suggested that the exploitation management reduce the 30 percent of the demand. One of the most important result of this will be the 26 percent reduction of the drop. By developing an optimization model and by imposing the necessary constraints on the critical regions, and transferring water from other parts, it seems that the trend of the drop will be controllable at a admissible level (less than 10 meters until 1390). Using the optimization model will make a change in the increasing trend of the drop and an improvement in the situation of the aquifer.