Contamination of urban water distribution systems (WDS) is a critical issue due to the number of victims that might be impacted in a short period of time. Any effective rapid emergency response plan for reducing the number of sick people or deaths among those drinking the polluted water requires a reliable forecast of the water contamination zoning map (CZM). The water CZM is a visual representation of the spread of contamination at the time of contamination detection. This study presents a novel methodology based on the rough set theory (RST) for real-time forecasting of the CZM caused by simultaneous multi-point contamination injection in WDS. Our proposed methodology consists of (i) a Monte Carlo simulation model to capture the uncertainties in a multi-point deliberate contamination, (ii) a numerical simulation model for simulating pipe flow, and (iii) a rough set-based technique for real-time CZM for a WDS equipped with a set of monitoring stations. The proposed methodology can be used for any type of random contamination of WDSs as well as emergencies in deliberate contamination of water distribution networks.

由于可能在短时间内影响到的受害者人数众多，城市供水系统的污染是一个关键问题。任何有效的减少饮用水污染者中的病人或死亡人数的有效快速应急响应计划，都需要对水污染分区图（CZM）进行可靠的预测。水CZM是污染物检测时污染物扩散的直观表示。这项研究提出了一种基于粗糙集理论（RST）的新颖方法，用于实时预测WDS中同时进行的多点污染注入引起的CZM。我们提出的方法包括：（i）蒙特卡洛模拟模型，用于捕获多点故意污染中的不确定性；（ii）用于模拟管道流量的数值模拟模型，（iii）一种基于粗糙集的实时CZM技术，用于配备了一组监视站的WDS。所提出的方法可用于任何形式的WDS随机污染以及水分配网络蓄意污染的紧急情况。