Fuzzy multi-objective simulation-optimization of stepped spillways considering flood uncertainty

The design of stepped spillways is a multi-objective optimization problem in which the uncertainty of parameters plays a key role in the stepped spillways’ design. A Fuzzy Transformation Method (FTM), as a first attempt, was employed to address the uncertainty through a multi-objective optimization of the stepped spillway. The proposed methodology comprises two parts. A numerical model is developed by FLOW3D^® and the Multi-Layer Perceptron artificial neural network (MLP-ANN) meta-model was trained and validated to substitute the numerical model. Then, the meta-model is coupled with a multi-objective optimization model to find the optimal design scenarios in each α-cut level. Finally, two multi-criteria decision-making (MCDM) models are applied to achieve the lower and upper bounds of optimum solutions. The application of proposed methodology in Jarreh dam, Iran, shows its capability in satisfying the objectives and also addressing the uncertainty in stepped spillways’ design.