Water distribution networks are being designed by traditional methods based on rules-of-thumb and personal experience of the designer. However, since there is no unique solution to any network design, namely there are various combinations of
pipes, pumps, tanks all of which satisfy the same pressure and velocity restrictions, it is most probable that the design performed by traditional techniques is not the optimum one.
This study deals how an optimization technique can be a useful tool for a designer during the design to find a solution. The method used within the study is the partial enumeration technique developed by Gessler. The technique is applied by a
commercially available software, i.e. WADISO SA. The study is focused on discrepancies between a network designed by traditional techniques and the same network designed by partial enumeration method. Attention is given to steps of enumeration, which are basically grouping of pipes, candidate pipe size and price function assignments, to demonstrate that the designers can control all the phases of optimization process. In this respect, special attention is given to price functions to show the effect of them on the result. The study also revealed that the cost of fitting materials cannot be included in the price function although it may have significant
effect in a system composed of closely located junctions.
The results obtained from this study are useful to show that although optimization methods do not provide a definite solution / partial enumeration method can assist designers to select the optimum system combination.
| Identifer | oai:union.ndltd.org:METU/oai:etd.lib.metu.edu.tr:http://etd.lib.metu.edu.tr/upload/12606816/index.pdf |
| Date | 01 December 2005 |
| Creators | Keles, Gultekin |
| Contributors | Merzi, Nuri |
| Publisher | METU |
| Source Sets | Middle East Technical Univ. |
| Language | English |
| Detected Language | English |
| Type | M.S. Thesis |
| Format | text/pdf |
| Rights | To liberate the content for public access |
Page generated in 0.002 seconds