Developing the Performance of Modern Methods Using Multi-Objective Optimization in Urban Runoff Control

In this study the Best Management Practices (BMPs) and Low Impact Development (LID) was employed to control quality and quantity of runoff due to urban flooding. The hydraulic and hydrological models of urban runoff were simulated by the Storm Water Management Model (SWMM) and combined with None-dominated Sorting Genetic Algorithm (NSGAII). Three objective function were defined included: minimize the cost of BMP (Investment) in dollar, minimization of runoff volume over the catchment in m3, and minimization of pollutant load contain TSS (Total Suspended Solid) and Pb in kg at external node (Karaj River). The urban catchment studied was part of the northeast of Karaj City namely Azimiyeh located in Alborz Province. In this research five types of BMP-LID structures were defined to NSGA-II with various types and areas and this algorithm combine types of BMP-LID with considering land use of the 67 sub-catchments and produce solutions. Finally, seven scenarios were defined to assess the NSGA-II. The results shown: that by applying the solution obtained by NSGAII in study area, quantity and quality of flood, decreased by 41% and 23%, respectively. Also, the use of BMP-LID structures in separated form only could provide one of objective functions and not an optimal choice. The use of BMP-LID structures in combination form with appropriate area and considering land use of sub-catchments can be having good performance compared to separated mode. As well as, higher cost does not result in more efficiency, especially in the flood quantity discussion.