Improved genetic algorithms for optimal design of drainage systems

Abstract

The optimization of existing sewer systems or making new systems is and will remain one of the key issues in drainage management in our society. The problem consists of minimization of a nonlinear cost function subjected to nonlinear constraints. To overcome the difficulties of the optimization of drainage systems, in this paper, an elitist adaptive genetic algorithm (EAGA) for pipe optimization has been developed, by integrating elitist genetic algorithm (EGA) with adaptive genetic algorithm (AGA). We compare the performance of the EAGA with that of EGA and AGA in optimizing sewer systems. The EAGA converges to the global optimum in far fewer generations than the EGA and AGA. We believe that the EAGA is the first step in realizing a class of self-organizing genetic algorithms (GAs) capable of adapting themselves in locating the global optimum in drainage systems.