Optimization of water network using big bang-big crunch algorithm

Recycling water in industrial plants has become necessary to minimize treatment cost and freshwater purchase expenses. Mathematical programming was used to solve water network optimization problems to find minimum water consumption or minimum wastewater generation. The purpose of this study was to develop a procedure to design water treatment and reuse networks using Erol and Eksins Big Bang-Big Crunch (BB-BC) algorithm. Another aim was to investigate the convergence characteristic and efficiency of the algorithm. Five case studies were used to test the performance of the BB-BC. The parameters that were varied were step size function, factor and penalty weights, resulting in fifteen configurations. Lowering the parameter value of the step size function was found to create near optimal and consistent solutions in all case studies. However, increasing the penalty function weight reduced convergence time while decreasing the factor has obtained low standard deviation. The configuration which has a step size value of 0.01, factor of 0.99 and a penalty weight of 100 was found to generate near optimal and consistent solutions in all case studies. This configuration has achieved the lowest convergence point at less than 200 iterations, lowest standard deviation and obtained the nearest flow rate value with the correction solution The BB-BC algorithm at configuration of step size = 0.01, = 0.99 and penalty weight = 100 also outperformed Pikaia, a public-domain genetic algorithm code, in all case studies