posted on 2017-09-01, 15:31authored byQiang Xu, Kuo Liu, Shunping Zhao, Nan Cao, Zhimin Qiang, Weiwei Ben
Water loss control is a key issue for most water supply companies all over the world because of freshwater scarcity and ever-rising water demand. Beijing is one of the cities facing severe water scarcity, so water loss control has been emphasized during the past decades. However, due to the huge size of the water distribution network, there is a great need to optimize the water loss control strategies. The water distribution network of Beijing has been planned to be partitioned into over 700 DMAs. Assessment of the DMAs’ water loss conditions and prediction of the achievements of different water loss control measures are critical to manage these DMAs. The goal of this paper is to develop a mathematical model to reveal how low the water loss of a DMA could go, so as to optimize the water loss control strategies. 36 DMAs were selected as study areas and data of the lowest minimal nigh flow (LMNF) and DMA characters (including pipe material, pipe length, number of properties, pipe age, and water pressure) were collected. The relationship between LMNF and DMA characters was established using multi-variant regression method. The model fit the data with R 2 =0.8. The model was then compared to the commonly used water loss indicator UARL and its sensitivity to the input variables was analyzed. Finally, the model was applied by Beijing Waterworks Group to optimize its water loss control strategies.