Extended Abstract
The set of factors including rainfall decrease, population increase, excessive use and waste of water in water supply networks has caused the world to face a serious shortage of water resources. Preventing water loss is one of the most important indicators of preserving water resources. In all sectors of water consumption (agriculture, drinking water, industry, etc.), water loss is an inevitable challenge. In urban water distribution networks, the high amount of water loss is due to their large size and multiple connections.
Chavar city located in Ilam province, Iran, is situated 15 kilometers northwest of the provincial capital. The population of this city is 10,554 people, of whom 5,831 live in the central part of Chavar city. Most of the city's distribution network pipes are made of asbestos, polyethylene pipes and even some metal pipes with small diameters. There are asbestos pipes with diameters of 50, 110, 100 and 200 mm, polyethylene pipes with diameters of 110 and 125 mm; and steel pipes with diameters of 25, 38 and 200 mm. With field surveys, the simulated water network was updated and the hydraulic analysis of the network was done by Auto CAD, GIS, Water GEMS software. According to the location of the subscribers, their consumption was allocated to the nearest pipe. In order to measure the pressure in different points of the network, oil type (glycerin) gauges were used. An ultrasonic portable flow meter was used to measure the flow rate.
In the first step of analysis of the transmission line, the flow rate was measured at the water supply well, the transmission pipeline and the two reservoirs in the transmission system. The design of the transmission line in two sections was modeled by Water GEMS software. The first part included a well to a concrete reservoir of 700 cubic meters, and the second part included a reservoir of 700 to a 1000 cubic meters. There were four pumps in this network. The PMP-1 pump belonged to the well, which had a flow rate of 56.29 liters per second and a pump head of 153.3 meters. PMP-2, PMP-3 and PMP-4 pumps were connected to the 700 cubic meter tank, which operated as two active pumps and one in standby mode during the day, and one active pump with two pumps in standby mode during the night. In the case of two active pumps, each pump had a flow rate of 38.44 liters per second and a head of 158.3 meters. The pipelines analyzed in the software showed that the first transmission line transferred a flow rate of 56.2 liters per second at a velocity of 1.79 meters per second to the 700 cubic meter reservoir. The second transmission line transferred the flow rate of 76.8 liters per second at a velocity of 1.22 meters per second for the 1000 cubic meter reservoir to the distribution network. The velocity in both pipelines was acceptable according to the range of design criteria, and the flow rate was completely consistent with the design flow rate of the pumps. The results of network simulation showed that 57% of the network nodes were in the peak consumption mode. This was due to the level difference of more than 130 meters between the reservoir and the consumer, as well as the absence of pressure reducing valves. Therefore, the pressure was outside the permissible range of 14 to 45 meters of water. Also, 73 percent of the pipelines lack the optimal velocity of 0.3 to 2.5 meters per second due to the inappropriate diameter of the pipelines. In the second simulation mode with 0.7 maximum flow rate (corresponding to the lowest amount of consumption in the coldest seasons of the year), the pressure in the nodes was improving, and 64.5% of the nodes had the optimal pressure, but the velocity remained outside the standard permissible range in 84% of the pipelines. The simulation in the transmission lines was validated by comparing the discharge field measurement using the ultrasonic flow meter and the value calculated in the software. The pressure of the network was measured at several points by the pressure gauge. Then these values were compared with the pressures obtained from the simulation in the lowest consumption mode. The results indicated that there was a good correlation between the hydraulic simulation results and the field evaluation of the network.