Detection of water-quality contamination events based on multi-sensor fusion using an extented Dempster–Shafer method

Abstract

This study presents a method for detecting contamination events of sources of drinking water based on the Dempster-Shafer (D-S) evidence theory. The detection method has the purpose of protecting water supply systems against accidental and intentional contamination events. This purpose is achieved by first predicting future water-quality parameters using an autoregressive (AR) model. The AR model predicts future water-quality parameters using recent measurements of these parameters made with automated (on-line) water-quality sensors. Next, a probabilistic method assigns probabilities to the time series of residuals formed by comparing predicted water-quality parameters with threshold values. Finally, the D-S fusion method searches for anomalous probabilities of the residuals and uses the result of that search to determine whether the current water quality is normal (that is, free of pollution) or contaminated. The D-S fusion method is extended and improved in this paper by weighted averaging of water-contamination evidence and by the analysis of the persistence of anomalous probabilities of water-quality parameters. The extended D-S fusion method makes determinations that have a high probability of being correct concerning whether or not a source of drinking water has been contaminated. This paper's method for detecting water-contamination events was tested with water-quality time series from automated (on-line) water quality sensors. In addition, a small-scale, experimental, water-pipe network was tested to detect water-contamination events. The two tests demonstrated that the extended D-S fusion method achieves a low false alarm rate and high probabilities of detecting water contamination events.

Adaptive Detection Method for Organic Contamination Events in Water Distribution Systems Using the UV-Vis Spectrum Based on Semi-Supervised Learning

“…Kühnert et al [4] conducted principal component analysis (PCA) to extract the eigenvalues of monitored water quality indicators and combined T2 statistics to serve as a base for detecting abnormal events. Hou et al [5] proposed a multifactor fusion algorithm for detecting water quality anomaly events based on autoregression and fuzzy C-means clustering. Their algorithm significantly improves detection performance under multi-index evidence conflicts.…”

Field Study of Real-Time Water Quality Control

“…The LCPF method uses the weighted average of previous data to estimate the value at the next step. The MVNN approach compares the Euclidean distance between the current measured water quality and any water quality measurement within the recent past in the multivariate space [9]. Once the value at the next step is available from the sensors, it is compared with the estimated one by calculating the residual (difference between observed and predicted value) in units of standard deviation.…”

“…A water contamination event is identified when the real-time water quality data are outside the expected range of allowable water quality criteria, at which point an alert is issued [9]. The main objective is to configure the EDS in a way to decrease the number of missed detection, with the minimum number of false alarms.…”

Detection of contaminants in water supply: A review on state-of-the-art monitoring technologies and their applications

“…Obtaining an early detection system using multiple sensor data station on-site is more beneficial compared to the data from a single site [53]. Most of the water quality parameters are used as primary indicators for contamination events in WDS, which are obtained from an online database, such as CANARY Database [54][55][56][57]. The main objective of the event detection method is to: (1) identify the possibility of event occurred, (2) characterize the event into subgroup (e.g., spatial area, time duration and severity level), and (3) detect contamination as accurately and early as possible.…”