Abstract
This study addresses the problem of identifying the source location of a contaminant spill in a river system when a sensor network returns observations containing random measurement errors. To solve this problem, we suggest a new framework comprising three main steps: (i) spill detection, (ii) data preprocessing, and (iii) source identification. Specifically, we applied a statistical process control chart to detect a contaminant spill with measurement errors while keeping the false alarm rate at less than or equal to a user-specified value. After detecting a spill, we generated a nonlinear regression model to estimate a breakthrough curve of the observations and derive a characteristic vector of the estimated curve. Using the characteristic vector as an input, a random forest model was constructed with the sensor raising the first alarm. The model provides output values between 0 and 1 to represent the possibility of each candidate location being the true spill source. These possibility values allow users to identify strong candidate locations for the spill. The accuracy of our framework was tested on part of the Altamaha River system in Georgia, USA.
Highlights
Water is a crucial resource for both public health and ecological life
We suggest a new framework comprised of three main steps: (i) detecting a contaminant spill, (ii) preprocessing obtained contaminant spill data, and (iii) identifying the spill source location via random forest models
This study proposed a new framework to identify the source location of a contaminant spill in a river
Summary
Water is a crucial resource for both public health and ecological life. Since the amount of fresh water is decreasing at the same time that population, industrialization, and environmental pollution are increasing, the importance of water quality monitoring is attracting more attention. Based on improvements to real-time sensor and data analysis technologies that enable people to monitor water quality more effectively, the problem of identifying the source location of a contaminant spill has been extensively studied by researchers. Most previous studies related to this problem have addressed two types of water systems: groundwater and rivers Optimization algorithms, such as linear/nonlinear programming and meta-heuristics, have been commonly used to identify contaminant spill locations in groundwater systems, as shown by Aral and Guan [1], Aral et al [2], Gorelick et al [3], Singh and Datta [4], and Sun et al [5]. An observation Xt y j is returned by the sensor installed at y j location, and the values from all K sensors at time t are represented by the observation vector: Xt ( y1 )
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call