Coupled Genetic Algorithm-LP Method For Single Contaminant Water Reuse Network Synthesis

Optimization of water networks for water using systems is inherently a mixed integer nonlinear programming (MINLP) problem. Many methods have been proposed for solving MINLP problems, that among these methods, deterministic algorithms assume convexity to insure convergence to global optimum solution, and a main difficulty of stochastic methods is its weakness in handing highly constrained problems. To overcome these weaknesses, the best solution has been chosen in this work, which is coupling genetic algorithm as a stochastic method and linear programming as a deterministic method. Linear programming has been chosen because of its simplicity by converting nonlinearities to linearities. Water using units in networks considered as a chromosome. Each chromosome consists of processes for generating different structures of a network, which are optimized with genetic algorithms, and then these structures are optimized by LP method.

Keywords:

water network , Water reuse , Optimization , Linear programming , Genetic algorithm