A Genetic Algorithm for the Minimum Cost Design of a Stormwater System

Abstract
Optimization methodology for design of stormwater systems is developed. The methodology uses a Genetic Algorithm Cost Minimization tool (GA-CM) to evaluate stormwater drainage system project costs. Also used are design capacity and water quality controls, real-world design standards, cost analysis, PCSWMM and version 4.4HGUX of the US-EPA SWMM program. It was successfully applied to a realistic but hypothetical stormwater system to select a near-optimal (minimum cost) set of design parameters.
The GA-CM considered standard design practices from (i) the Ministry of the Environment of Ontario 2003 Stormwater Management Practices Planning and Design Manual and (ii) design information collected from interviews with consultants. The detail provided in the GA-CM is perhaps beyond what consultants feel that they need today. Interviews with consultants emphasized the need to address sizing of significant design parameters (e.g. depth of storage facility). This was the focus of the optimization methodology developed.
Genetic algorithm routines are a powerful tool for the selection of the best combination of stormwater system design parameters. Semi-automatic optimization of urban drainage systems and associated costs will lead to improved urban drainage design practices and improve stormwater quality discharges to downstream receiving waters.
This paper is only available in PDF Format: View full text PDF