Identifying Contaminant Intrusion in Water Distribution Networks under Water Flow and Sensor Report Time Uncertainties

Contamination events in water distribution networks (WDNs) could have severe health and economic consequences. Contaminants can be deliberately or accidentally introduced into the WDN. Quick identification of the injection location and time is important in devising a mitigation plan to prevent further spread of the contaminant in the network. A method of identifying the possible intrusion point in a given network and reporting data is to use an inverse calculation by backtracking the potential path of the contaminant in the network. However, there is an element of uncertainty in the data used for calculation, particularly in water flow and sensor report time. Given the uncertainties, a method was developed in this study for fast and accurate contaminant source identification. This paper proposes a comparison filter of results by first identifying potential contaminant locations through backtracking, followed by a forward calculation to determine the injection time range, thereby reducing the potential suspects and providing likeliness comparison among the suspects. The effectiveness of the proposed method was examined by applying it to a benchmark WDN. By simulating uncertainties and filtering through the results, several possible contaminant intrusion locations and times were identified.

Interdependence of flow and pipe characteristics in transient induced contamination intrusion: numerical analysis

Contaminant intrusion in pipelines during transients is a remarkable mechanism, which leads to a decline in the quality of the contained water. The negative pressure of water hammer pressure waves is the trigger for the suction of pollution from the surrounding leak area, and hence deteriorating water quality. The volume of contamination intruded into the pipeline is investigated using mathematical and numerical modeling of the phenomenon. To elucidate this phenomenon in real pipe systems, the intrusion amount is estimated for 72 different scenarios including: two lengths of pipeline (i.e. short and long), three different leak locations, three different fluid velocities in the pipe, two leak diameters and two pipeline materials (elastic and viscoelastic). The results showed that the amount of intrusion in viscoelastic pipes was clearly less than that in elastic pipes, especially in long pipelines. The critical zone of high intrusion risk is identified close to the downstream valve for small leak sizes, nevertheless, it is difficult to estimate this zone in the case of large leaks due to significant interactions between nodal components (valve, leak, reservoir).

Experimental Investigations and CFD Modeling of Contaminant Intrusion in a Water Network during Transient Events

Intrusion in drinking water networks (DWN) can be triggered by transient low-pressure events. This intrusion may result in the contamination of drinking water supplied to consumers, which may have major health impacts. This research aims to investigate the influence of a DWN’s operating parameters on the intrusion and progression of the contaminant in a pilot drinking water network setup. Results show that the minimum time required for the contaminant to begin entering the system was influenced by the system operating pressure. Faster initial intrusion times were observed under low operating pressure. In addition, the crack size influenced the time required for the contaminant to fully intrude the system. Similarly, the time required for the contaminant to reach certain points in the DWN was influenced by the operating pressure. These results were verified using two additional tools; a high-speed camera was used to monitor the contaminant transient progression through the DWN under different pressures; and computational fluid dynamics modeling was used to calculate the corresponding contaminant concentration. The results suggest that the ability of the system to quickly stabilize its operating pressure would reduce the probability of a contaminant intrusion into the DWN.

Contaminant intrusion in water distribution networks: review and proposal of an integrated model for decision making

Contaminant intrusion in a distribution network (DN) refers to the entry of harmful chemicals and pathogens in the presence of three conditions: (i) the availability of a contaminant source near water mains; (ii) a pathway: leakage or breakage; and (iii) a driving force: low or negative pressure in the water main. The occurrence of contamination in a DN can take place frequently as there is no specific treatment at this stage except secondary disinfection. Contaminant intrusion requires as much attention as source water protection or treatment plants, particularly given that at this point, water is near the final stage prior to human consumption. Failure to detect and treat at this time could have potential negative impacts on consumers’ health. Following the September 11, 2001 attack, strict regulations are now enforced by the municipalities to monitor water quality within DNs. This review article focuses on various aspects of contaminant intrusion in DNs based on more than 90 journal articles, peer-reviewed conference proceedings, and research reports. Here we present details on the conditions of contaminant intrusion, water quality regulations, sampling, protection and mitigation strategies, and various modelling approaches for decision making. Based on this review, we propose an integrated model that will help guide effective decision making for contaminant detection and mitigation.