مهدی خراشادی زاده

Pollution of surface water especially rivers has created serious problems for both human health and environment. Finding efficient solutions for controlling river pollution is essential in order to reduce damage for the consumer to protect the environment. In this research, river quality management planning is determined in a way that minimized the rate of pollution damage with using particle swarm optimization. The case study is Gheshlagh River in Iran, Kurdistan province. The timetable for entering contamination is at different hours of the day, and eight source points enter contamination to the Gheshlagh River. Also in all hours of the day, water harvesting from the Gheshlagh River is carried out by the downstream consumers. The 192 of decision variables of this research are corresponding to the amount of contamination entering from these sources at different times. Also, 8 decision variables are corresponding to the optimal location of contamination entering to the river.  The results indicate that in the optimal condition the value of contamination does not exceed from the allowable limited.  However in the unmanaged condition, the contamination value exceeds the allowable limit. The river quality management planning which determines the optimal location and the timetable of pollution discharge, reduced significantly consumer's damages. So that the objective function of this study was reduced by 97.7%.

Understanding of the fate of temporal and spatial distribution of pollutionis an essential subject for prediction of damages, caused by pollution, on the ecology of rivers and coastal areas. It is also necessary to provide efficient solutions for both pollution control and environmental protection. In this study, shallow water equations have been used to simulate pollution transport by 2-dimensional finite volume method. Calibration-modification and frequent changing of the amount of parameters is a known issue in hydraulic and hydrologic models. Therefore, it is necessary to utilize methods for sensitivity analysis and reduction of the parameters number to calibrate models. In this study, the RSA sensitivity analysis method was used for each parameter in which the ratio of sensitivity and cumulative distribution function for the sets of good and bad parameters are computed.. For this purpose, 5000 iterations from uncertainty domain of calibration parameters of pollution transport model in a Standard issue of shallow water were performed by using uncertainty algorithm GLUE. With enforcing the acceptable threshold values for the sumof square error index on the total simulation results, 1000 premier simulations were introduced as an efficient simulation. The corresponded set of parameters was considered as good set parameters and the others as bad set parameters. Thus, the sensitivity index was calculated for the Manning coefficient and the floor slope in the x and y directions. The comparison of sensitivity analysis of parameters based on the RSA methods and variation coefficient of parameters indicated that RSA is an efficient method for sensitivity analysis of model parameters.

In shallow river¡ regardless of the vertical acceleration against gravity acceleration¡ great simplifying will be created in equations that the result of them are shallow water equations. In this studyshallow water equation has been used to simulate transport of pollution in 2D limited volumes method. In this method it is possible to turn differential equations ruling fluid to be solved in numerical form. To achieve better results in estimating pollution concentration in the river and Reliable range¡ the pollution Transformation Model uncertainty analysis prepared were examined. To uncertainty assessment in this study 5000 times repetition of uncertainty 3 calibration parameter of pollution transferring model by the use of uncertainty algorithm GLUE had happened in river. For this matter of pollution Transformation Problem breaking a dam was used. By applying acceptability threshold indicator of total average cubes (SSE) on all the obtained simulations¡ 1000 top simulations were known as efficient simulations. Considering about 95 percent confidence as high and low bound of uncertainty¡ and by applying used d-factor and p-factor indicators¡ 68 and 75 respectively were obtained Which reflects the high level of concentration was observed in 95 percent confidence. Results had showed using GLUE method caused reduction of uncertainty pollution transferring model parameters. Extraction probabilistical charts last corresponding parameter distributed with efficient simulation¡ parameter (n) as a sensitive and effective parameter on simulation models. And the optimal value obtained for this parameter 0/2477.