Optimal Design of Storm Sewer Network Based on Risk Analysis by Combining Genetic Algorithm and SWMM Model

The design of an urban storm sewer network is a costly task. Therefore, the design should be done so that the total cost becomes minimal. This requires modeling the problem in an optimization form. Floods are stochastic. Designing such a system is associated with risk. Thus, a project is optimal when both design costs and potential future risks are incorporated. This means that the selection of rainfall-runoff return period has to be based on risk analysis. SWMM software was used to handle hydraulic network simulation and the Network optimization was performed using a genetic algorithm in which the decision variables were the diameter and slope of the pipes. To calculate the cost of runoff damage, relationships for land uses, infrastructure and traffic, were provided. The accuracy of the simulation- optimization model seeking the optimal design of the storm sewer network was approved by a benchmark network evaluation. The developed model was implemented in a region of Tehran city to determine the optimal design return period with a risk analysis approach. The results showed that the 10-year return period with is the optimal return period with an annual damage risk cost of 508.68 billion rials, an annual design cost of 943.78 billion rials and a total cost of 1452.45 billion rials. Therefore, the developed method in which the genetic algorithm and SWMM model are combined in addition to the risk-based design approach is an effective tool for the optimal design of storm sewer networks.