Developing a Model for Optimizing the Geometric Characteristics of Water Diversion Systems

Abstract

Determining dimensions of a dam diversion system has always been a challenge for designers because of uncertainties regarding flood characteristics and acceptable risk to the diversion system. This study presents a numerical model to determine optimal features of a diversion system. The optimization model uses a genetic algorithm to determine optimal dimensions of a water diversion system based on the construction costs, failure costs, and acceptable risk to the diversion system by considering hydraulic uncertainties and flood risk analysis. The objective function of the optimization model aims to minimize construction costs and expected value of damages caused by floods with varying return periods. Karun-IV Dam located in southwest Iran is considered as a case study. The results show that the expected annual average cost of the diversion system (capital cost + expected damage) is lower than the consultant’s estimated figure. In this study, a spillway is added to handle the maximum observed flood in order to match the consultant’s reliability with the model’s reliability. The results indicate that even with the added spillway, the optimal system is more effective than the premier plan suggested by the project consultant. Therefore, investigating flood risk and its uncertainty could provide a better framework for the design and cost for the configuration of diversion systems in dams.