Optimal Design of Open Channel Sections Using PSO Algorithm

Abstract

This paper applies an evolutionary algorithm, the particle swarm optimization (PSO), to design the optimum open channel section. Depth, channel side slope and bottom width are considered as the variables for rectangular, triangular and trapezoidal channels, respectively. The objective function is minimizing the construction cost of the channel section. MATLAB software is used for programming and doing the optimization process. Manning's uniform flow formula has been used as a constraint for the optimization model. The cost function is included the cost of earthwork, the increment in the cost of earthwork with the depth below the ground surface and the cost of lining. Simple functions of unit cost terms have been used to express the optimum values of section variables. The optimum section variables are obtained for the case of minimum area or minimum wetted perimeter problems. The results of this study showed that the PSO is a robust algorithm to compute the optimum section variables in open channel design.