Eau de Paris (EdP) swiftly equipped a leading edge Command- and Control System (CCS) for network operations and -supervision purposes.
From the beginning of 2010, EdP is the single operator for the production and distribution of water for Paris city. This enabled the company to integrate information into the CCS from an additional hundred flow meters that are dedicated to district metering areas (DMA) that were previously operated by other distribution operators. Since the former distribution operators deployed automatic meter reading (AMR) on the 93.000 customer’s meters in Paris in 2009, the information from these meters is now transmitted to the CCS once a day. All this information is compiled and analysed to detect leakages in the network.
In order to achieve and maintain a constant water pressure in the network, this has been divided into sub-networks depending on ground elevation. These sub-networks are separated by valves. The pressure in the sub-networks is controlled by remote control valves and by a SCADA system including 15.000 information points, 200 flow meter points and 120 pressure levels. Since the flow in the constant-pressurised sub-networks is related to actual consumption, the information captured allows shaping typical consumption curves that are used for leakage detection. By real-time monitoring of the pressure in a sub-network, a large drop in pressure would indicate a pipe break. The real-time flow is also compared with the typical consumption curves shaped with historical data. The detection sensitivity of this method is limited to important breaks (up to 500 m3/h). Progressive leakage is detected by short differed time detection tools. The water distributed in the network is analysed from a macroscopic vision towards a microscopic vision, from networks to sub-networks. Daily consumption in these different zones is calculated from flow meters. Usually all the areas present similar trends. Increasing difference between two linked areas indicates an abnormality which can either be result from leakage or exceptional consumption like the refilling of a municipal swimming pool. These analyses allow the detection of small anomalies such as progressive leakage.
The next step, which is currently being implemented, is to compare the inflow in an area with the outflow measured by the buildings meters and transmitted by AMR. Differences between these two curves can only be explained by consumption not measured by AMR, free authorized consumption or leakage either on public or private networks.