Abstract:
Several consequences can arise from unexpected terrorist actions such as an attack by chemical into the drinking water distribution system. To investigate possible threats of chemical contamination in the system to detect early events and actions can help water distribution network operators making assertive decisions. This work presents a model capable of event detection parathion contamination from the monitoring of quality sensors in the network. The methodology considers the chemical reactions of the parathion pesticide when present in water into the problem the detection, using the program EPANET-MSX. The data generated from the hydraulic model and the drinking water quality served to create a pattern recognition system using the artificial neural network NARX in order to, estimate the behavior of chlorine in the network, while a change point identification methodology is used to identify significant changes caused by the contaminant inserted in the network. An estimated error analysis was performed and simulated attacks were identified.
Reference:
OLIVEIRA, Eva Carolline M.; BRETAN, Bruno Melo; DANTAS, Renato F.; MACEDO, Letícia dos Santos; Luvizotto Junior, Edevar. Detection of chemical intrusion compounds in water distribution networks by quality sensors data mining. In: INTERNATIONAL JOINT CONFERENCE IN WATER DISTRIBUTION SYSTEMS ANALYSIS AND COMPUTING AND CONTROL IN THE WATER INDUSTRY, 1., 2018, Kingston, Ontario Canada. Proceedings… 8 p.
Access to the article on the Event website:
https://ojs.library.queensu.ca/index.php/wdsa-ccw/article/view/12141/7734
Several consequences can arise from unexpected terrorist actions such as an attack by chemical into the drinking water distribution system. To investigate possible threats of chemical contamination in the system to detect early events and actions can help water distribution network operators making assertive decisions. This work presents a model capable of event detection parathion contamination from the monitoring of quality sensors in the network. The methodology considers the chemical reactions of the parathion pesticide when present in water into the problem the detection, using the program EPANET-MSX. The data generated from the hydraulic model and the drinking water quality served to create a pattern recognition system using the artificial neural network NARX in order to, estimate the behavior of chlorine in the network, while a change point identification methodology is used to identify significant changes caused by the contaminant inserted in the network. An estimated error analysis was performed and simulated attacks were identified.
Reference:
OLIVEIRA, Eva Carolline M.; BRETAN, Bruno Melo; DANTAS, Renato F.; MACEDO, Letícia dos Santos; Luvizotto Junior, Edevar. Detection of chemical intrusion compounds in water distribution networks by quality sensors data mining. In: INTERNATIONAL JOINT CONFERENCE IN WATER DISTRIBUTION SYSTEMS ANALYSIS AND COMPUTING AND CONTROL IN THE WATER INDUSTRY, 1., 2018, Kingston, Ontario Canada. Proceedings… 8 p.
Access to the article on the Event website:
https://ojs.library.queensu.ca/index.php/wdsa-ccw/article/view/12141/7734
