| dc.description.abstract | The water distribution system conveys water from the water treatment plant (WTP) to the user. A significant amount of the water is lost in the transit. The International Water Supply Association (IWSA) claims that the amount is typically 20-30% of the total water production depending on the overall state of the distribution system (Hunaidi et al., 2000). There is a great variation of the pipe material and the quality of construction of trenches for pipelines. Therefore, some distribution systems may lose as much as 50%. IWSA made a survey, which stated that the major cause of unaccounted for water loss is leakage.
Oslo Water and Sewerage Works (VAV) wants to investigate options to optimize the efficiency of water used, and of the water demand strategies, to increase the reliability of the service provided. The motivation for this is that future projections for water demand in Oslo resulting from population growth are considered unacceptable. As such, the future goal of VAV is to maintain a stable and constant water production close to 100 mill. m3/year in the coming years regardless of the population growth (Hovedplan vannforsyning 2015-2030, 2015). This can be achieved by reducing specific water consumption and by reducing water leakages in the network from about 30% to less than 20%. In order to achieve these goals, there is a need to promote new thinking and smart innovations for the water services.
This master thesis will specifically focus on analysing the specifications of a supervisory control and data aquisition (SCADA) system to be able to retrieve consistent data, in the right format, in order to calibrate hydraulic network simulation models (HNSMs). Among the main results from the thesis is the establishment of a link between a centralized SCADA database and a HNSM. Useful insights in data extraction will be provided together with an example of how the data can be used in model refinement. A pilot system of the water supply in Oslo, called the E3-network, has been used as an example.
The HNSM used has been refined by implementing new demand patterns calculated from actual, retrieved data from a data historian. Also, the β-value of the model has been refined in order for the model to be more accurate. | |
