Hydraulic-based optimization algorithm for the design of stormwater drainage networks

Abstract

Stormwater drainage networks are designed to reduce the risk of rainwater damage to the served area. The purpose of optimizing a stormwater drainage system is to reduce overall construction costs and to meet hydraulic design requirements. Currently, designs that rely on software or manual calculations are limited by the available time and the designer’s capabilities. In fact, manual optimization for large networks consumes a lot of time and effort, and there is no guarantee that the optimal design is reached, also it is subject to human errors. In recent years, several researchers have focused on creating optimization design algorithms specifically for sewer and storm networks, such as genetic algorithm (GA), linear programming (LP), heuristic programming (HP),…etc. However, these studies were limited to covering one or two design parameters and constraints. Additionally, in some studies, the hydraulic performance of the designed network was not treated in a proper way, especially the water surface profile effects. So, the main objective of the study is to develop an effective hydraulic-based optimization algorithm (HBOA) that can dynamically get the optimal design with minimum total cost for a given storm network layout and meet all hydraulic requirements. To achieve this, a MATLAB code is created and coupled with SewerGEMS software that automatically simulates all expected optimization scenarios based on network hydraulic performance. The HBOA is validated economically and hydraulically using two benchmark examples from the literature. According to the economic validation, the total network cost generated by HBOA was the lowest when compared to the optimization methods found in the literature. During the hydraulic evaluation, it was observed that the optimization algorithm (GA-HP) used in the literature for the benchmark examples does not meet the hydraulic requirements where the networks are flooded, whereas HBOA meets the hydraulic requirements with minimal overall network cost. Also, the HBOA is applied to four real stormwater drainage networks that were already designed, constructed, and optimized manually. The four redesigned real cases using HBOA revealed a cost reduction of about 15% compared to the original designs, while consuming a few hours for the design and optimization processes. Finally, the developed HBOA is a robust, time-efficient, and cost-effective optimization and hydraulic design tool which could be used in the design of stormwater drainage networks with different design constraints with minimal human interference.