Integrating hydraulic and gis modeling for assessing water losses Case study; Kumasi Southeast District

Thumbnail Image
April, 2016
Journal Title
Journal ISSN
Volume Title
There are several challenges facing water utility companies. One of such challenge is high level of water losses (non-revenue water) in a distribution network. Non-revenue water (NRW) is water that has been produced but cannot be billed. The loss can be as a result of leakages (real losses), theft of water (apparent losses) and free use (unbilled authorized consumption). In the Ashanti Region of Ghana, Non-revenue water (NRW) levels have been consistently high. This has affected revenue and quality of service to customers. The main objective of this research was to use an integration of a hydraulic model and GIS to estimate non-revenue water in Kumasi Southeast District. In determining the magnitude of non-revenue water, the International Water Association methodology and empirical flow rates were used. The hydraulic modeling software ‘Epanet’ was used in modeling the water distribution network. In modeling the pipe network, the Hazen-Williams algorithm for water flow rate, friction and headloss was used. Parameters such as pressure, flow, unit headloss, velocity and friction factor were obtained for the various pipes and nodes upon a successful run of the model. The pressure values generated by the hydraulic and the GIS model at each node were used to determine the background leakages at the various nodes. This was done using the standard hydraulic equation. GIS was used in modeling the water distribution network. The research shows that non-revenue water level was as high as 49% of the total water supplied to Southeast District of Ghana Water Company Limited (GWCL). The 49% of the system input volume (SIV) represented 283,146m3 of treated water, out of which 215,068m3 was lost through real losses. This represents 75.96% of the non-revenue water. Apparent losses contributed 67,948m3, representing 24% of the total non-revenue water. Unbilled authorized consumption also contributed a loss of 129m3 of the system input volume (SIV). This represents 0.04% of the total non-revenue water. The model also revealed the level of background leakages at each node. The highest level of background leakage was 0.087669m3 while the lowest level of background leakage was 0.001854m3. The study concluded that hydraulic and GIS models are complementary technologies and their integration provided access to more reliable, up-to-date information and reduces response time to tackle water losses.
A Thesis submitted to The Department of Geomatic Engineering, College of Engineering In Partial Fulfillment of the Requirement for the Degree of Master of Science, Geomatic Engineering April, 2016