Determination the optimal dimensions of concrete gravity dam by using metaheuristic algorithms (Comparison of algorithms)

Concrete gravity dams secure their stability by the weight of the concrete used in their structure. Therefore, minimizing their weight (the volume of concrete consumed in their body) can reduce the costs significantly. This study aims to evaluate the performance of three metaheuristic optimization algorithms: harmony search, particle swarm optimization, and artificial bee colony, to find the optimal cross-section size of the gravity dam. In this way, the Koyna dam located in India is considered a case study. The programming is applied in Matlab software. Each algorithm under the constraints of this problem (the sliding, overturning, and vertical tension on the body of the dam) is run 6 times. Finally, the lowest value was chosen as the optimal result. The results revealed that however all the algorithms have the optimal outputs than their real one but the optimum one is for the harmony search algorithm. To investigate the role of available uncertainties of dam cross-section, Monte Carlo simulation is engaged. The achieved results based on reliability show more safety of dam design.