{"id":30,"date":"2019-12-03T13:02:32","date_gmt":"2019-12-03T13:02:32","guid":{"rendered":"http:\/\/www.kios.ucy.ac.cy\/battledim\/?page_id=30"},"modified":"2020-01-22T11:35:40","modified_gmt":"2020-01-22T11:35:40","slug":"challenge","status":"publish","type":"page","link":"https:\/\/battledim.ucy.ac.cy\/?page_id=30","title":{"rendered":"Challenge"},"content":{"rendered":"\n

L-Town (see Fig. 1) is a small hypothetical town with a population of around 10,000 people. The water utility of L-Town is responsible for delivering drinking water to consumers through a network of pipes with a total length of 42.6km. In previous years, the utility was experiencing a large number of pipe breaks and water losses, affecting its service quality. The water distribution network of L-Town is receiving water from two (2) reservoirs, and the water utility aims at providing water with a pressure head of at least 20m to all of its consumers. A Pressure Reduction Valve (PRV) is installed in the lower part of the town (\u201cArea B\u201d), to help reducing background leakages. PRVs are also installed downstream of the two main reservoirs, to help regulating the pressure. A pump and a water tank have been installed in the higher part of the town (\u201cArea C\u201d), to provide sufficient pressure to the consumers of that area. The tank has a diameter of 16 meters with a cylindrical shape. The pump has been programmed so that the tank should be refilling during the night and emptying to \u201cArea C\u201d during the day. There are three consumer types in L-Town: residential, commercial and industrial. During workdays (Monday to Friday), water consumption follows a similar pattern, whereas during the weekend (Saturday and Sunday), there is higher consumption during late hours as the result of night life. Areas with industrial users do not follow the same pattern of consumption. L-Town is located in the Northern hemisphere, thus higher water usage is expected around July\/August, and lower in December\/January. For the purposes of this challenge, there are no significant variations of water consumption during holidays or other special days.<\/p>\n\n\n\n

\"The
Fig.1: The L-Town Benchmark Network<\/figcaption><\/figure><\/div>\n\n\n\n

To enhance\nits capability in monitoring water losses, the water utility of L-Town decided\nto install one (1) tank water level sensor, three (3) flow sensors, and 33 pressure\nsensors (see Fig. 2), all transmitting their measurements every 5 minutes to\nthe utility\u2019s Supervisory Control and Data Acquisition (SCADA) System. Pressure\nsensors give an average value of the last 5 minutes, which mitigates the\nuncertainty due to pressure transients in the system. In addition, 82 Automated Metered Readings (AMRs)\nhave been installed in \u201cArea C\u201d, for delivering water consumption data directly\nto the SCADA system (see Fig. 3). Each AMR gives the aggregated consumption of\nmany users in the AMR area.<\/p>\n\n\n\n

During 2018,\na number of leakage events occurred, which were detected and fixed by the water\nutility. However, it is believed that a number of smaller leakages occurred but\nnot revealed. It is also assumed that some leakages occurred abruptly, whereas\nothers developed gradually, as incipient events, from background leaks into\npipe bursts. <\/p>\n\n\n\n

To assist the\nL-Town water utility decision-making process, the utility developed an\nEPANET-based nominal model of the distribution network, in which base demands\nwere assigned to nodes, following historical data of proximity consumers.\nMoreover, two nominal demand patterns were identified for residential and\ncommercial consumer types (with some discrepancies). The utility believes that\nthere might be some inaccuracies in the model, e.g., with respect to the pipe\nroughness and pipe diameters. In addition, the utility was not able to confirm\nthe status of all the valves in the network (i.e., whether they are open or\nclosed).<\/p>\n\n\n\n

The L-Town\nwater utility is searching for a solution to help them analyzing the SCADA\ndataset, and detect leakage events as fast as possible. In addition, it is crucial\nfor the utility to have an indication where approximately the leakage occurs,\nso that the field workers can inspect those potential leaks using their\nequipment. <\/p>\n\n\n\n

\"The
Fig. 2: The locations of the 33 pressure sensors placed in L-Town are indicated with yellow color along with their node ID.<\/figcaption><\/figure><\/div>\n\n\n\n
\"82
Fig.3: 82 AMR locations (nodes with red color) in \u201cArea C\u201d of L-Town. Nodes colored in blue do not have AMRs installed.<\/figcaption><\/figure><\/div>\n\n\n\n

Competition Goal<\/h2>\n\n\n\n

The L-Town\nutility has created an open call for teams to demonstrate their ability in\ndetecting and localizing leakage events. The teams will be given a historical\nSCADA dataset along with information related with the leakages detected and fixed\nby the utility throughout 2018, to use for training purposes and for\ncalibrating their models. It is possible that more leakage events occurred during\n2018, however the utility was not able to detect and localize them. <\/p>\n\n\n\n

Throughout 2019,\nthe utility conducted periodic surveys using additional sensing equipment, pipe\ninspections and other methods, and was able to detect and isolate all the\nleakage events that occurred within that period. The most critical of these\nevents were repaired, however it was not possible to repair some of these leakages\ndue to financial reasons. <\/p>\n\n\n\n

The overall\ngoal of this competition, is to identify methods which are able to detect and\nlocalize the leakage events that occurred in L-Town in 2019, as fast as\npossible (with respect to time) and as accurately as possible (with respect to\ntheir location), in order to minimize their overall financial costs, both in\nwater losses, as well as due to the hours spent in isolating the leakage by the\nutility staff. The L-Town utility will compare the different solutions and\nselect the best one based on that objective. More details regarding the\nevaluation procedure will be released in January 2020.<\/p>\n","protected":false},"excerpt":{"rendered":"

L-Town (see Fig. 1) is a small hypothetical town with a population of around 10,000 people. The water utility of L-Town is responsible for delivering drinking water to consumers through a network of pipes with a total length of 42.6km. In previous years, the utility was experiencing a large number of pipe breaks and water … Continue reading “Challenge”<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"templates\/template-full-width.php","meta":{"footnotes":""},"class_list":["post-30","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=\/wp\/v2\/pages\/30","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=30"}],"version-history":[{"count":8,"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=\/wp\/v2\/pages\/30\/revisions"}],"predecessor-version":[{"id":129,"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=\/wp\/v2\/pages\/30\/revisions\/129"}],"wp:attachment":[{"href":"https:\/\/battledim.ucy.ac.cy\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=30"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}