Closure to “Knowledge-Based Optimization Model for Control Valve Locations in Water Distribution Networks” by Mohammed E. Ali

Abstract

The writer is extremely thankful to the discusser, and the following is prepared as a response. The objective of the paper was to reduce excess pressure within the distribution system and the associated background leakage. This is done by introducing flow control vales at various strategic locations within the system. The paper acknowledges that the method in most situations will not create an isolated pressure zone or district metering area (DMA) as it was not the objective. If background leakage can be best minimized by creating a pressure zone, the valve setting will be zero (to simulate a shut valve) because the setting varies between zero (closed valve) and one (fully open valve). The creation of DMAs will not reduce leakage by itself, but it is the first step toward a better system control. The objective in establishing DMAs in a distribution system is to identify the economic level of real losses within each DMA, and then measures to reduce this can easily be designed and implemented (MacDonald and Yates 2005). Also, creation of DMAs has an important role in leakage management because abnormal flow measurement at the inlet can highlight the presence of a leak within the DMA (Grayman et al. 2009). Once these DMAs are created, excess pressure can then be minimized to reduce the system background leakage. The method developed in the paper attempts to reduce leakage within zones whether they are pressure zone or a strategic supply network by minimizing unwanted pressure. Therefore, the model developed is timeless and should benefit other researchers in the field. DMAs can have various shapes and sizes and can be very complex. Reducing of excess pressure in such DMAs cannot be done by simply introducing a control valve at the DMA inlet. Therefore, formal optimization to locate and set control valves within the system or subsystem is still of great importance to pressure management and background leakage reduction. Real systems have redundancy built within them when they are initially designed not only to ensure the security of uninterrupted supply during peak demand but also to cater to future or unexpected demand. Although built-in redundancy improves robustness of the system, it leads to higher pressure within a pressure zone or a DMA and hence a higher associated background leakage. By introducing control valves at various strategic locations within the system, it will lead to a reduced background leakage and pressures that are sufficient to meet the system pressure requirement (Germanopoulos and Jowitt 1989). The objectives of published research in the automatic creation of DMAs do not consider leakage directly as their primary concern. This discussion is opening new possibilities for the study of leakage reduction in pipe networks by incorporating this at the stage of DMA design. The writer is aware of the simplistic nature of the Jowitt and Xu (1990) example network, and that the size of the system used is small when compared with real systems. For comparison reasons and availability of data, this example network was used, similarly to other researchers who have published work in this area. The writer appreciates the comments and observations made.