Maximum entropy design of water distribution systems under multiple operating conditions

This paper presents an optimal design methodology for the design of water distribution systems based on a multi-objective genetic algorithm, namely NSGA II, coupled with water distribution network simulation software, EPANET 2 and subroutine that calculates entropy. The novelty of this research is that it is a reliability-based approach that combines least cost design based on real-world commercial pipe sizes with maximum entropy for water distribution systems working under multiple operating conditions. The new approach has been demonstrated by designing a well known benchmark network. Results generated using multiple operating conditions and the commonly used single operating condition are presented and compared. It is shown that designs based on multiple operating conditions do not entail excessive cost, have more uniform pipe diameters and are more reliable.