Keywords: pipe failure, diameter resizing, total energy loss, node pressure control, node to node pipe path grid, net pipe failure index.

The design of water supply network systems tries to achieve simultaneously different aims related to the desired performance of the hydraulic system. The criteria to define this performance is presently still mostly based on empirical and subjective criteria, giving a poorly documented support to justify the technical decision.

The system conception starts with the definition of the topology of the network system. For reasons of flexibility concerning pipe failure situations looped network configurations are usually chosen.

For hydraulic design, the decision criteria currently used are: (a) minimum cost; (b) maximum flexibility and; (c) minimum pressure service values (correlated with minimum head losses).

In practical situations engineers must choose and justify a technical solution that must be a compromise between the objective functions previously defined.

The objective decision criteria, in water supply network systems design, becomes clearer and easier to understand if presented in a numerical and, or graphical language. So, one of the challenges is to achieve improvement in objective decision criteria is the definition of numerical indicators that may quantify suggestively the level of performance of the system.

In the cases studied and presented in this paper the following numerical indicators were evaluated: (a) total cost (CR); (b) entropy of the looped network (SR); (c) head loss (ER); (d) net pipe failure index (NPFI) [1] and; (e) flow in the pipe (QR). The cases studied were prepared to analyze the influence of: (a) reservoir location and; (b) the number of pipes diverging from the source node.

The cases studied (Source node-number of Loops-number of Nodes-number of Pipes) are (S1-L2-N6-P7), (S1-L3-N6-P8), (S3-L2-N6-P7).

The first hydraulic design of the looped networks was based on the maximum entropy flow distribution in the pipes.

To achieve the aim of having a prefixed minimum value of the pressure in all nodes of the looped network, even in a situation of single pipe failure, it is obvious that the diameter of the pipes must increase. Pipe diameter resizing was performed with an iterative procedure [1].

To identify, during calculations, the total energy loss between the source node and all the other nodes of the system, it was defined, for each looped network, a binary "node to node pipe path grid",

The results obtained are presented in both tables and figures.

1

Malafaya-Baptista, M., "Water Supply Looped Systems Design. Pipe Failure and Pressure Distributions. Pressure Restrictions and Design Procedures", CCWI2005, Water Management for the 21st Century, Exeter, United Kingdom, 2005.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £105 +P&P)