Application of ant colony optimization to optimal design of open channels

Artificial open channels are expensive infrastructure in terms of material, construction and maintenance and application of optimization models has a great role in minimizing their cost. Two optimization approaches for the cross-sectional design are compared. In the first stage, an equivalent roughness coefficient is considered without any sectional division according to the Manning equation. Then, in the second stage, the horizontal velocity variation over the cross-section is taken into account by dividing the channel into three segments. Both, ant colony optimization, as a new heuristic approach, and the genetic algorithm method have been used to solve the non-linear optimization problem. Considering the obtained results for different design variables, the optimal and appropriate cross-section is proposed for composite trapezoidal open channels by the second model by considering horizontal velocity variation. Furthermore, the efficiency and capability of the ant colony optimization are examined to solve and analyze the optimal design models.