Performance Improved Multi-Objective Optimization in Applying Low-Impact Development Strategies to Control Urban Runoff

Best Management Practices (BMPs) can play a vital role to control natural disasters like floods. In this paper, retention pond and vegetative swale are considered to restrain urban runoff. Storm water management modeling (SWMM) is used for runoff modeling. A piece of code is developed based on Non-dominated Sorting Genetic Algorithm (NSGA-II) in MATLAB to optimize the BMPs application. The aim is comparing the effect of roulette wheel, tournament and random selection operators to obtain the optimal location and area of BMPs. Minimizing the runoff volume and pollution in sub-catchments and the construction cost of the BMPs are three objective functions. Rafsanjan city located in southeast of Iran is selected as an appropriate case study. Estimating the best pressure of selection operator in roulette wheel and the best selection size in tournament operator and simultaneous quantitative and qualitative optimization using two BMPs are the innovations of this study. The results indicate that the pressure of the selection operator in roulette wheel which leads to the optimal answer is three and nine while the best size of selection in the tournament operator is nine. Optimum location, type, area and volume for each BMP are obtained after running the code.