Multiple-contaminant water network synthesis

Abstract

Water is the main resources for process industry. Many industrial companies pay very high cost in each year for the water usage. The saving on water usage decreases the operating and capital costs of company. Water network synthesis can help reduce the water usage. This work uses General Algebraic Modelling System program (GAMS) for generating the water network system with multiple contaminants by using Mixed-Integer non-liner programming model (MINLP) with objective function minimizing fresh water usage and wastewater. The initialization step starts with non-liner programming (NLP) to estimate the initial water flow rate before final design step minimizing fresh water. This work uses data from the literature of Savelski et al. (2003) and Koppol et al. (2003) as a case study of water/wastewater network to compare with the result of our study. Our water network with multiple contaminants is represented as grid diagrams and process flow diagram using lower fresh water usage than one from case study process.