Development of an Unconditional Mathematical Model to Design Sewer Networks

Design of sewer networks involves many constraints and technical criteria. These issues make the problem complicated and the feasible alternatives difficult to achieve. Besides, the methods of handling these constraints play a major role in optimization approaches for the least cost design. In this work an integrated analysis-design model is developed thereby all constraints are automatically satisfied during designing. A normal design alternative is defined herein which is a vector of values between zero and one. This vector represents the sewers diameters, slopes and pump station locations. The normal alternative is then decoded to real design parameters based on the hydraulics rules and the problem constraints. Using the proposed model every normal alternative is feasible and consequently the input design parameters are totally unconditional. At the end the abilities of the model is shown introducing and solving a relatively large sewer network. Furthermore, to demonstrate the capability and easiness of integrating the model with optimization tools, a Simulated Annealing (SA) algorithm is also applied to obtain the least cost design of the example. The results show that the optimum design and the rate of optimization convergence are significantly improved using the proposed method.