A structural-based model for water network synthesis

[Display omitted] •Systematic decomposition method for water network synthesis is presented.•Method is developed for direct reuse/recycle and regeneration networks.•Model accounts for water flowrates/cost and piping numbers/cost.•Pipeline numbers and costs decrease with more clusters formed in the water network.•More fresh water is consumed with the decrease of network complexity. This paper presents a novel systematic decomposition method for the synthesis of water network. The main objective of this approach is to reduce the complexity of the water network by dividing the system into several clusters with optimisation constraints. These constraints include lengths of pipeline, total pipe lengths of the clusters, etc. The method was applied to both direct reuse/recycle and water regeneration networks. A literature example is used to demonstrate the novel approach. The generated network alternatives were compared for their minimum fresh water and wastewater flowrates, piping costs, as well as the number of pipelines. It is observed that the number of pipelines and the piping costs decrease when the system is divided into more clusters (i.e. simpler network). The results also show that when the complexity of the network decreases, higher fresh water is observed for the direct reuse/recycle case, and higher regenerated flowrate for the case of water regeneration network. Both cases lead to higher operating and total annualised cost of the water networks. Hence, there is a trade-off between network complexity and water flowrates. It is up to the plant engineers or designers to decide the level of complexity and water saving that they can accept.