A dynamic thresholds scheme for contaminant event detection in water distribution systems

Classification machine learning to detect de facto reuse and cyanobacteria at a drinking water intake

Harmful algal blooms (HABs) or higher levels of de facto water reuse (DFR) can increase the levels of certain contaminants at drinking water intakes. Therefore, the goal of this study was to use multi-class supervised machine learning (SML) classification with data collected from six online instruments measuring fourteen total water quality parameters to detect cyanobacteria (corresponding to approximately 950 cells/mL, 2900 cells/mL, and 8600 cells/mL) or DFR (0.5, 1 and 2 % of wastewater effluent) events in the raw water entering an intake. Among 56 screened models from the caret package in R, four (mda, LogitBoost, bagFDAGCV, and xgbTree) were selected for optimization. mda had the greatest testing set accuracy, 98.09 %, after optimization with 7 false alerts. Some of the most important water parameters for the different models were phycocyanin-like fluorescence, UVA254, and pH. SML could detect algae blending events (estimated <9000 cells/mL) due in part to the phycocyanin-like fluorescence sensor. UVA254 helped identify higher concentrations of DFR. These results show that multi-class SML classification could be used at drinking water intakes in conjunction with online instrumentation to detect and differentiate HABs and DFR events. This could be used to create alert systems for the water utilities at the intake, rather than the finished water, so any adjustment to the treatment process could be implemented.

Gated graph neural networks for identifying contamination sources in water distribution systems

Contamination events in water distribution networks (WDN) pose significant threats to water supply and public health. Rapid and accurate contamination source identification (CSI) can facilitate the development of remedial measures to reduce impacts. Though many machine learning (ML) methods have been proposed for fast detection, there is a critical need for approaches capturing complex spatial dynamics in WDNs to enhance prediction accuracy. This study proposes a gated graph neural network (GGNN) for CSI in the WDN, incorporating both spatiotemporal water quality data and flow directionality between network nodes. Evaluated across various contamination scenarios, the GGNN demonstrates high prediction accuracy even with limited sensor coverage. Notably, directional connections significantly enhance the GGNN CSI accuracy, underscoring the importance of network topology and flow dynamics in ML-based WDN CSI approaches. Specifically, the method achieves a 92.27% accuracy in narrowing the contamination source to 5 points using just 2 h of sensor data. The GGNN showcases resilience under model and measurement uncertainties, reaffirming its potential for real-time implementation in practice. Moreover, our findings highlight the impact of sensor sampling frequency and measurement accuracy on CSI accuracy, offering practical insights for ML methods in water network applications.

Generative adversarial networks for detecting contamination events in water distribution systems using multi-parameter, multi-site water quality monitoring

Contamination events in water distribution networks (WDNs) can have a huge impact on water supply and public health; increasingly, online water quality sensors are deployed for real-time detection of contamination events. Machine learning has been used to integrate multivariate time series water quality data at multiple stations for contamination detection; however, accurate extraction of spatial features in water quality signals remains challenging. This study proposed a contamination detection method based on generative adversarial networks (GANs). The GAN model was constructed to simultaneously consider the spatial correlation between sensor locations and temporal information of water quality indicators. The model consists of two networks-a generator and a discriminator-the outputs of which are used to measure the degree of abnormality of water quality data at each time step, referred to as the anomaly score. Bayesian sequential analysis is used to update the likelihood of event occurrence based on the anomaly scores. Alarms are then generated from the fusion of single-site and multi-site models. The proposed method was tested on a WDN for various contamination events with different characteristics. Results showed high detection performance by the proposed GAN method compared with the minimum volume ellipsoid benchmark method for various contamination amplitudes. Additionally, the GAN method achieved high accuracy for various contamination events with different amplitudes and numbers of anomalous water quality parameters, and water quality data from different sensor stations, highlighting its robustness and potential for practical application to real-time contamination events.

The role of deep learning in urban water management: A critical review

Deep learning techniques and algorithms are emerging as a disruptive technology with the potential to transform global economies, environments and societies. They have been applied to planning and management problems of urban water systems in general, however, there is lack of a systematic review of the current state of deep learning applications and an examination of potential directions where deep learning can contribute to solving urban water challenges. Here we provide such a review, covering water demand forecasting, leakage and contamination detection, sewer defect assessment, wastewater system state prediction, asset monitoring and urban flooding. We find that the application of deep learning techniques is still at an early stage as most studies used benchmark networks, synthetic data, laboratory or pilot systems to test the performance of deep learning methods with no practical adoption reported. Leakage detection is perhaps at the forefront of receiving practical implementation into day-to-day operation and management of urban water systems, compared with other problems reviewed. Five research challenges, i.e., data privacy, algorithmic development, explainability and trustworthiness, multi-agent systems and digital twins, are identified as key areas to advance the application and implementation of deep learning in urban water management. Future research and application of deep learning systems are expected to drive urban water systems towards high intelligence and autonomy. We hope this review will inspire research and development that can harness the power of deep learning to help achieve sustainable water management and digitalise the water sector across the world.

Developing stacking ensemble models for multivariate contamination detection in water distribution systems

This study presents a new stacking ensemble model for contamination event detection using multiple water quality parameters. The stacking model consists of a number of machine learning base predictors and a meta-predictor, and it is trained using cross-validation to capture different features in multiple water quality parameters and then used for water quality predictions. For each water quality parameter, the residuals between predicted and measured data are classified to identify anomalies with thresholds derived from the sequential model-based optimization method and detection probabilities updated using Bayesian analysis. Alarms derived from individual water quality parameters are fused to enhance the anomaly signals and improve the detection accuracy. The proposed stacking-based method is evaluated using a data set of six water quality parameters from a real water distribution system with randomly simulated events. The stacking-based method could detect 2496 events out of a total 2500 events without a false alarm. The results show that the stacking method outperforms an artificial neural network (ANN) benchmark method in contamination event detection. The stacking method has a higher true positive rate, lower false positive rate and higher F1 score than the ANN method. This implies that the stacking method has great promise of detecting contamination events in the water distribution system.

Using data mining techniques to isolate chemical intrusion in water distribution systems

The security of water distribution systems has become the subject of an increasing volume of research over the last decade. Data analysis and machine learning are linked to hydraulic and quality modeling for improving the capacity of water utilities to save lives when faced with the contamination of water networks. This research applies k-nearest neighbor and random forest algorithms to estimate the location of contamination sources at near-real time. Epanet and Epanet-MSX software are used to simulate intrusions of pesticide into water distribution system and the interaction with compounds already present in water bulk. Different pesticide concentrations are considered in the simulations, and chlorine monitoring occurs through placed quality sensors. The results show that random forest can localize [Formula: see text] of contamination scenarios, while the KNN algorithm found [Formula: see text]. Finally, an assessment of contamination spread is made for a better understanding of the impacts of non-localized contamination.

Two novelty learning models developed based on deep cascade forest to address the environmental imbalanced issues: A case study of drinking water quality prediction

Environmental quality data sets are typically imbalanced, because environmental pollution events are rarely observed in daily life. Prediction of imbalanced data sets is a major challenge in machine learning. Our recent work has shown deep cascade forest (DCF), as a base learning model, is promising to be recommended for environmental quality prediction. Although some traditional models were improved by introducing the cost matrix, little is known about whether cost matrix could enhance the prediction performance of DCF. Additionally, feature extraction is also an important way to potentially improve the model's ability to predict the imbalanced data. Here, we developed two novelty learning models based on DCF: cost-sensitive DCF (CS-DCF) and DCF that combines unsupervised learning models and greedy methods (USM-DCF-G). Subsequently, CS-DCF and USM-DCF-G were successfully verified by an imbalanced drinking water quality data set. Our data presented both CS-DCF and USM-DCF-G show better prediction performance than that of DCF alone did. In particular, USM-DCF-G shows the best performance with the highest F1-score (95.12 ± 2.56%), after feature extraction and selection by using unsupervised learning models and greedy methods. Thus, the two learning models, especially USM-DCF-G, were promising learning models to address environmental imbalanced issues and accurately predict environmental quality.

Fish forewarning of comprehensive toxicity in water environment based on Bayesian sequential method

Environmental impact of pollutants can be analyzed effectively by acquiring fish behavioral signals in water with biological behavior sensors. However, a variety of factors, such as the complexity of biological organisms themselves, the device error and the environmental noise, may compromise the accuracy and timeliness of model predictions. The current methods lack prior knowledge about the fish behavioral signals corresponding to characteristic pollutants, and in the event of a pollutant invasion, the fish behavioral signals are poorly discriminated. Therefore, we propose a novel method based on Bayesian sequential, which utilizes multi-channel prior knowledge to calculate the outlier sequence based on wavelet feature followed by calculating the anomaly probability of observed values. Furthermore, the relationship between the anomaly probability and toxicity is analyzed in order to achieve forewarning effectively. At last, our algorithm for fish toxicity detection is verified by integrating the data on laboratory acceptance of characteristic pollutants. The results show that only one false positive occurred in the six experiments, the present algorithm is effective in suppressing false positives and negatives, which increases the reliability of toxicity detections, and thereby has certain applicability and universality in engineering applications.

Machine learning in natural and engineered water systems

Water resources of desired quality and quantity are the foundation for human survival and sustainable development. To better protect the water environment and conserve water resources, efficient water management, purification, and transportation are of critical importance. In recent years, machine learning (ML) has exhibited its practicability, reliability, and high efficiency in numerous applications; furthermore, it has solved conventional and emerging problems in both natural and engineered water systems. For example, ML can predict various water quality indicators in situ and real-time by considering the complex interactions among water-related variables. ML approaches can also solve emerging pollution problems with proven rules or universal mechanisms summarized from the related research. Moreover, by applying image recognition technology to analyze the relationships between image information and physicochemical properties of the research object, ML can effectively identify and characterize specific contaminants. In view of the bright prospects of ML, this review comprehensively summarizes the development of ML applications in natural and engineered water systems. First, the concept and modeling steps of ML are briefly introduced, including data preparation, algorithm selection and model evaluation. In addition, comprehensive applications of ML in recent studies, including predicting water quality, mapping groundwater contaminants, classifying water resources, tracing contaminant sources, and evaluating pollutant toxicity in natural water systems, as well as modeling treatment techniques, assisting characterization analysis, purifying and distributing drinking water, and collecting and treating sewage water in engineered water systems, are summarized. Finally, the advantages and disadvantages of commonly used algorithms are analyzed according to their structures and mechanisms, and recommendations on the selection of ML algorithms for different studies, as well as prospects on the application and development of ML in water science are proposed. This review provides references for solving a wider range of water-related problems and brings further insights into the intelligent development of water science.

Using machine learning classification to detect simulated increases of de facto reuse and urban stormwater surges in surface water

Water quality events such as increases in stormwater or wastewater effluent in drinking water sources pose hazards to drinking water consumers. Stormwater and wastewater effluent enter Lake Mead-an important drinking water source in the southwest USA-via the Las Vegas Wash. Previous studies have applied machine learning and online instruments to detect contamination in water distribution systems. However, alert systems at drinking water intakes would provide more time for corrective action. An array of online instruments measuring pH, conductivity, redox potential, turbidity, temperature, tryptophan-like fluorescence, UV absorbance (UVA₂₅₄), TOC, and chlorophyll-a was fed raw water directly from Lake Mead. Wastewater effluent, dry weather Las Vegas Wash, and storm-impacted Las Vegas Wash samples were blended into the instrument inlets at known ratios to simulate three types of adverse water quality events. Data preprocessing was conducted to correct for diurnal patterns or instrument drift. Supervised machine learning was conducted using previously published models in R. Ninety-nine models were screened on the raw data. Eight high-performing models were evaluated in-depth and optimized. Weighted k-Nearest Neighbors, Single C5.0 Ruleset, Mixture Discriminant Analysis, and an ensemble of these three models had accuracy over 97% when assigning test set data among three classes (Normal, Event, or Maintenance). The ensemble detected all event types at the earliest timepoint and had one false positive that was not a lag error (i.e., consecutively following a true positive). Omitting Maintenance, the Adaboost model had over 99% test set accuracy and zero false positives that were not lag errors. Data preprocessing was beneficial, but the optimal methods were model-specific. All nine water quality variables were useful for most models, but UVA₂₅₄ and turbidity were most important.

Review of Modeling Methodologies for Managing Water Distribution Security

Water distribution systems are vulnerable to hazards that threaten water delivery, water quality, and physical and cybernetic infrastructure. Water utilities and managers are responsible for assessing and preparing for these hazards, and researchers have developed a range of computational frameworks to explore and identify strategies for what-if scenarios. This manuscript conducts a review of the literature to report on the state of the art in modeling methodologies that have been developed to support the security of water distribution systems. First, the major activities outlined in the emergency management framework are reviewed; the activities include risk assessment, mitigation, emergency preparedness, response, and recovery. Simulation approaches and prototype software tools are reviewed that have been developed by government agencies and researchers for assessing and mitigating four threat modes, including contamination events, physical destruction, interconnected infrastructure cascading failures, and cybernetic attacks. Modeling tools are mapped to emergency management activities, and an analysis of the research is conducted to group studies based on methodologies that are used and developed to support emergency management activities. Recommendations are made for research needs that will contribute to the enhancement of the security of water distribution systems.

Water characterization and early contamination detection in highly varying stochastic background water, based on Machine Learning methodology for processing real-time UV-Spectrophotometry

Water is a resource that affects every aspect of life. Intentional (terrorist or wartime events) or accidental water contamination events could have a tremendous impact on public health, behavior and morale. Quick detection of such events can mitigate their effects and reduce the potential damage. Continuous on-line monitoring is the first line of defense for reducing contamination associated damage. One of the available tools for such detection is the UV-absorbance spectrophotometry, where the absorbance spectra are compared against a set of normal and contaminated water fingerprints. However, as there are many factors at play that affect this comparison, it is an elusive and tedious task. Further, the comparison against a set of known fingerprints is futile when the water in the supply system are a mix, with varying proportions, of water from different sources, which differ significantly in their physicochemical characteristics. This study presents a new scheme for early detection of contamination events through UV absorbance under unknown routine conditions. The detection mechanism is based on a new affinity measure, Fitness, and a procedure similar to Gram based amplification, which result in a flexible mechanism to alert if a contamination is present. The method was shown to be most effective when applied to a set of comprehensive experiments, which examined the absorbance of various contaminants in drinking water in lab and real-life configurations. Four datasets, which contained real readings from either laboratory experiments or monitoring station of an operational water supply system were used. To extend the testbed even further, an artificial dataset, simulating a vast array of proportions between specific water sources is also presented. The results show, that for all datasets, high detection rates, while maintaining low levels of false alarms, were obtained by the algorithm.

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

A method that uses the ultraviolet-visible (UV-Vis) spectrum to detect organic contamination events in water distribution systems exhibits the advantages of rapid detection, low cost, and no need for reagents. The speed, accuracy, and comprehensive analysis of such a method meet the requirements for online water quality monitoring. However, the UV-Vis spectrum is easily disturbed by environmental factors that cause fluctuations of the spectrum and result in false alarms. This study proposes an adaptive method for detecting organic contamination events in water distribution systems that uses the UV-Vis spectrum based on a semi-supervised learning model. This method modifies the baseline using dynamic orthogonal projection correction and adjusts the support vector regression model in real time. Thus, an adaptive online anomaly detection model that maximizes the use of unlabeled data is obtained. Experimental results demonstrate that the proposed method is adaptive to baseline drift and exhibits good performance in detecting organic contamination events in water distribution systems.

Model-based approach for cyber-physical attack detection in water distribution systems

Modern Water Distribution Systems (WDSs) are often controlled by Supervisory Control and Data Acquisition (SCADA) systems and Programmable Logic Controllers (PLCs) which manage their operation and maintain a reliable water supply. As such, and with the cyber layer becoming a central component of WDS operations, these systems are at a greater risk of being subjected to cyberattacks. This paper offers a model-based methodology based on a detailed hydraulic understanding of WDSs combined with an anomaly detection algorithm for the identification of complex cyberattacks that cannot be fully identified by hydraulically based rules alone. The results show that the proposed algorithm is capable of achieving the best-known performance when tested on the data published in the BATtle of the Attack Detection ALgorithms (BATADAL) competition (http://www.batadal.net).

Modeling marine oily wastewater treatment by a probabilistic agent-based approach

This study developed a novel probabilistic agent-based approach for modeling of marine oily wastewater treatment processes. It begins first by constructing a probability-based agent simulation model, followed by a global sensitivity analysis and a genetic algorithm-based calibration. The proposed modeling approach was tested through a case study of the removal of naphthalene from marine oily wastewater using UV irradiation. The removal of naphthalene was described by an agent-based simulation model using 8 types of agents and 11 reactions. Each reaction was governed by a probability parameter to determine its occurrence. The modeling results showed that the root mean square errors between modeled and observed removal rates were 8.73 and 11.03% for calibration and validation runs, respectively. Reaction competition was analyzed by comparing agent-based reaction probabilities, while agents' heterogeneity was visualized by plotting their real-time spatial distribution, showing a strong potential for reactor design and process optimization.

Contamination Event Detection with Multivariate Time-Series Data in Agricultural Water Monitoring

Time series data of multiple water quality parameters are obtained from the water sensor networks deployed in the agricultural water supply network. The accurate and efficient detection and warning of contamination events to prevent pollution from spreading is one of the most important issues when pollution occurs. In order to comprehensively reduce the event detection deviation, a spatial–temporal-based event detection approach with multivariate time-series data for water quality monitoring (M-STED) was proposed. The M-STED approach includes three parts. The first part is that M-STED adopts a Rule K algorithm to select backbone nodes as the nodes in the CDS, and forward the sensed data of multiple water parameters. The second part is to determine the state of each backbone node with back propagation neural network models and the sequential Bayesian analysis in the current timestamp. The third part is to establish a spatial model with Bayesian networks to estimate the state of the backbones in the next timestamp and trace the “outlier” node to its neighborhoods to detect a contamination event. The experimental results indicate that the average detection rate is more than 80% with M-STED and the false detection rate is lower than 9%, respectively. The M-STED approach can improve the rate of detection by about 40% and reduce the false alarm rate by about 45%, compared with the event detection with a single water parameter algorithm, S-STED. Moreover, the proposed M-STED can exhibit better performance in terms of detection delay and scalability.

Short-term forecasting of turbidity in trunk main networks

Water discolouration is an increasingly important and expensive issue due to rising customer expectations, tighter regulatory demands and ageing Water Distribution Systems (WDSs) in the UK and abroad. This paper presents a new turbidity forecasting methodology capable of aiding operational staff and enabling proactive management strategies. The turbidity forecasting methodology developed here is completely data-driven and does not require hydraulic or water quality network model that is expensive to build and maintain. The methodology is tested and verified on a real trunk main network with observed turbidity measurement data. Results obtained show that the methodology can detect if discolouration material is mobilised, estimate if sufficient turbidity will be generated to exceed a preselected threshold and approximate how long the material will take to reach the downstream meter. Classification based forecasts of turbidity can be reliably made up to 5 h ahead although at the expense of increased false alarm rates. The methodology presented here could be used as an early warning system that can enable a multitude of cost beneficial proactive management strategies to be implemented as an alternative to expensive trunk mains cleaning programs.

Real-time detection of organic contamination events in water distribution systems by principal components analysis of ultraviolet spectral data

The detection of organic contaminants in water distribution systems is essential to protect public health from potential harmful compounds resulting from accidental spills or intentional releases. Existing methods for detecting organic contaminants are based on quantitative analyses such as chemical testing and gas/liquid chromatography, which are time- and reagent-consuming and involve costly maintenance. This study proposes a novel procedure based on discrete wavelet transform and principal component analysis for detecting organic contamination events from ultraviolet spectral data. Firstly, the spectrum of each observation is transformed using discrete wavelet with a coiflet mother wavelet to capture the abrupt change along the wavelength. Principal component analysis is then employed to approximate the spectra based on capture and fusion features. The significant value of Hotelling's T² statistics is calculated and used to detect outliers. An alarm of contamination event is triggered by sequential Bayesian analysis when the outliers appear continuously in several observations. The effectiveness of the proposed procedure is tested on-line using a pilot-scale setup and experimental data.

Integrating physically based simulators with Event Detection Systems: Multi-site detection approach

The Fault Detection (FD) Problem in control theory concerns of monitoring a system to identify when a fault has occurred. Two approaches can be distinguished for the FD: Signal processing based FD and Model-based FD. The former concerns of developing algorithms to directly infer faults from sensors' readings, while the latter uses a simulation model of the real-system to analyze the discrepancy between sensors' readings and expected values from the simulation model. Most contamination Event Detection Systems (EDSs) for water distribution systems have followed the signal processing based FD, which relies on analyzing the signals from monitoring stations independently of each other, rather than evaluating all stations simultaneously within an integrated network. In this study, we show that a model-based EDS which utilizes a physically based water quality and hydraulics simulation models, can outperform the signal processing based EDS. We also show that the model-based EDS can facilitate the development of a Multi-Site EDS (MSEDS), which analyzes the data from all the monitoring stations simultaneously within an integrated network. The advantage of the joint analysis in the MSEDS is expressed by increased detection accuracy (higher true positive alarms and fewer false alarms) and shorter detection time.

Online flow cytometry reveals microbial dynamics influenced by concurrent natural and operational events in groundwater used for drinking water treatment

Detailed measurements of physical, chemical and biological dynamics in groundwater are key to understanding the important processes in place and their influence on water quality – particularly when used for drinking water. Measuring temporal bacterial dynamics at high frequency is challenging due to the limitations in automation of sampling and detection of the conventional, cultivation-based microbial methods. In this study, fully automated online flow cytometry was applied in a groundwater system for the first time in order to monitor microbial dynamics in a groundwater extraction well. Measurements of bacterial concentrations every 15 minutes during 14 days revealed both aperiodic and periodic dynamics that could not be detected previously, resulting in total cell concentration (TCC) fluctuations between 120 and 280 cells μL⁻¹. The aperiodic dynamic was linked to river water contamination following precipitation events, while the (diurnal) periodic dynamic was attributed to changes in hydrological conditions as a consequence of intermittent groundwater extraction. Based on the high number of measurements, the two patterns could be disentangled and quantified separately. This study i) increases the understanding of system performance, ii) helps to optimize monitoring strategies, and iii) opens the possibility for more sophisticated (quantitative) microbial risk assessment of drinking water treatment systems.

A canonical correlation analysis based method for contamination event detection in water sources

In this study, a general framework integrating a data-driven estimation model is employed for contamination event detection in water sources. Sequential canonical correlation coefficients are updated in the model using multivariate water quality time series. The proposed method utilizes canonical correlation analysis for studying the interplay between two sets of water quality parameters. The model is assessed by precision, recall and F-measure. The proposed method is tested using data from a laboratory contaminant injection experiment. The proposed method could detect a contamination event 1 minute after the introduction of 1.600 mg l(-1) acrylamide solution. With optimized parameter values, the proposed method can correctly detect 97.50% of all contamination events with no false alarms. The robustness of the proposed method can be explained using the Bauer-Fike theorem.

Why conventional detection methods fail in identifying the existence of contamination events

Early warning systems are widely used to safeguard water security, but their effectiveness has raised many questions. To understand why conventional detection methods fail to identify contamination events, this study evaluates the performance of three contamination detection methods using data from a real contamination accident and two artificial datasets constructed using a widely applied contamination data construction approach. Results show that the Pearson correlation Euclidean distance (PE) based detection method performs better for real contamination incidents, while the Euclidean distance method (MED) and linear prediction filter (LPF) method are more suitable for detecting sudden spike-like variation. This analysis revealed why the conventional MED and LPF methods failed to identify existence of contamination events. The analysis also revealed that the widely used contamination data construction approach is misleading.

Use of log-transformed absorbance spectra for online monitoring of the reactivity of natural organic matter

This study examined the significance of water quality monitoring parameters obtained via logarithmic transformation of the absorbance spectra of raw and treated drinking water. The data were generated using samples of the influent, settled and filtered water acquired weekly over a six months period at two full scale treatment plants. Examination of the weekly plant samples combined with the data of laboratory fractionation and chlorination experiments showed that the slopes of the log-transformed spectra are correlated with typically reported water quality parameters (e.g., its specific absorbance at 254 nm, SUVA254); yet the determination of spectral slopes is considerably simpler and potentially information-rich. The spectral slopes determined for the range of wavelength 280-350 nm were shown to be correlated with the yields of trihalomethanes (THMs) and haloacetic acids (HAAs). These results support the notion that multi-wavelength monitoring of the absorbance spectra of drinking water and their interpretation via logarithmic transformation constitutes a promising practically implementable approach for online water quality monitoring.

Network hydraulics inclusion in water quality event detection using multiple sensor stations data

Event detection is one of the current most challenging topics in water distribution systems analysis: how regular on-line hydraulic (e.g., pressure, flow) and water quality (e.g., pH, residual chlorine, turbidity) measurements at different network locations can be efficiently utilized to detect water quality contamination events. This study describes an integrated event detection model which combines multiple sensor stations data with network hydraulics. To date event detection modelling is likely limited to single sensor station location and dataset. Single sensor station models are detached from network hydraulics insights and as a result might be significantly exposed to false positive alarms. This work is aimed at decreasing this limitation through integrating local and spatial hydraulic data understanding into an event detection model. The spatial analysis complements the local event detection effort through discovering events with lower signatures by exploring the sensors mutual hydraulic influences. The unique contribution of this study is in incorporating hydraulic simulation information into the overall event detection process of spatially distributed sensors. The methodology is demonstrated on two example applications using base runs and sensitivity analyses. Results show a clear advantage of the suggested model over single-sensor event detection schemes.

A multivariate based event detection method and performance comparison with two baseline methods

Early warning systems have been widely deployed to protect water systems from accidental and intentional contamination events. Conventional detection algorithms are often criticized for having high false positive rates and low true positive rates. This mainly stems from the inability of these methods to determine whether variation in sensor measurements is caused by equipment noise or the presence of contamination. This paper presents a new detection method that identifies the existence of contamination by comparing Euclidean distances of correlation indicators, which are derived from the correlation coefficients of multiple water quality sensors. The performance of the proposed method was evaluated using data from a contaminant injection experiment and compared with two baseline detection methods. The results show that the proposed method can differentiate between fluctuations caused by equipment noise and those due to the presence of contamination. It yielded higher possibility of detection and a lower false alarm rate than the two baseline methods. With optimized parameter values, the proposed method can correctly detect 95% of all contamination events with a 2% false alarm rate.

An integrated logit model for contamination event detection in water distribution systems

The problem of contamination event detection in water distribution systems has become one of the most challenging research topics in water distribution systems analysis. Current attempts for event detection utilize a variety of approaches including statistical, heuristics, machine learning, and optimization methods. Several existing event detection systems share a common feature in which alarms are obtained separately for each of the water quality indicators. Unifying those single alarms from different indicators is usually performed by means of simple heuristics. A salient feature of the current developed approach is using a statistically oriented model for discrete choice prediction which is estimated using the maximum likelihood method for integrating the single alarms. The discrete choice model is jointly calibrated with other components of the event detection system framework in a training data set using genetic algorithms. The fusing process of each indicator probabilities, which is left out of focus in many existing event detection system models, is confirmed to be a crucial part of the system which could be modelled by exploiting a discrete choice model for improving its performance. The developed methodology is tested on real water quality data, showing improved performances in decreasing the number of false positive alarms and in its ability to detect events with higher probabilities, compared to previous studies.

A method of detecting contamination events using multiple conventional water quality sensors

Early warning systems are often used for detecting contamination accidents. Traditional event detection methods suffer from high false negative and false positive errors. This paper proposes a detection method using multiple conventional water quality sensors and introduces a method to determine the values of parameters, which was configured as a multiple optimization problem and solved using a non-dominated sorting genetic algorithm (NSGA-II). The capability of the proposed method to detect contamination events caused by cadmium nitrate is demonstrated in this paper. The performance of the proposed method to detect events caused by different concentrations was also investigated. Results show that, after calibration, the proposed method can detect a contamination event 1 min after addition of cadmium nitrate at the concentration of 0.008 mg/l and has low false negative and positive rates.

Integrated hydraulic and organophosphate pesticide injection simulations for enhancing event detection in water distribution systems

As a complementary step towards solving the general event detection problem of water distribution systems, injection of the organophosphate pesticides, chlorpyrifos (CP) and parathion (PA), were simulated at various locations within example networks and hydraulic parameters were calculated over 24-h duration. The uniqueness of this study is that the chemical reactions and byproducts of the contaminants' oxidation were also simulated, as well as other indicative water quality parameters such as alkalinity, acidity, pH and the total concentration of free chlorine species. The information on the change in water quality parameters induced by the contaminant injection may facilitate on-line detection of an actual event involving this specific substance and pave the way to development of a generic methodology for detecting events involving introduction of pesticides into water distribution systems. Simulation of the contaminant injection was performed at several nodes within two different networks. For each injection, concentrations of the relevant contaminants' mother and daughter species, free chlorine species and water quality parameters, were simulated at nodes downstream of the injection location. The results indicate that injection of these substances can be detected at certain conditions by a very rapid drop in Cl2, functioning as the indicative parameter, as well as a drop in alkalinity concentration and a small decrease in pH, both functioning as supporting parameters, whose usage may reduce false positive alarms.

Contamination event detection using multiple types of conventional water quality sensors in source water

Early warning systems are often used to detect deliberate and accidental contamination events in a water system. Conventional methods normally detect a contamination event by comparing the predicted and observed water quality values from one sensor. This paper proposes a new method for event detection by exploring the correlative relationships between multiple types of conventional water quality sensors. The performance of the proposed method was evaluated using data from contaminant injection experiments in a laboratory. Results from these experiments demonstrated the correlative responses of multiple types of sensors. It was observed that the proposed method could detect a contamination event 9 minutes after the introduction of lead nitrate solution with a concentration of 0.01 mg L(-1). The proposed method employs three parameters. Their impact on the detection performance was also analyzed. The initial analysis showed that the correlative response is contaminant-specific, which implies that it can be utilized not only for contamination detection, but also for contaminant identification.

A coupled classification - evolutionary optimization model for contamination event detection in water distribution systems

This study describes a decision support system, alerts for contamination events in water distribution systems. The developed model comprises a weighted support vector machine (SVM) for the detection of outliers, and a following sequence analysis for the classification of contamination events. The contribution of this study is an improvement of contamination events detection ability and a multi-dimensional analysis of the data, differing from the parallel one-dimensional analysis conducted so far. The multivariate analysis examines the relationships between water quality parameters and detects changes in their mutual patterns. The weights of the SVM model accomplish two goals: blurring the difference between sizes of the two classes' data sets (as there are much more normal/regular than event time measurements), and adhering the time factor attribute by a time decay coefficient, ascribing higher importance to recent observations when classifying a time step measurement. All model parameters were determined by data driven optimization so the calibration of the model was completely autonomic. The model was trained and tested on a real water distribution system (WDS) data set with randomly simulated events superimposed on the original measurements. The model is prominent in its ability to detect events that were only partly expressed in the data (i.e., affecting only some of the measured parameters). The model showed high accuracy and better detection ability as compared to previous modeling attempts of contamination event detection.

Controllability analysis as a pre-selection method for sensor placement in water distribution systems

Detection of contamination events in water distribution systems is a crucial task for maintaining water security. Online monitoring is considered as the most cost-effective technology to protect against the impacts of contaminant intrusions. Optimization methods for sensor placement enable automated sensor layout design based on hydraulic and water quality simulation. However, this approach results in an excessive computational burden. In this paper we outline the application of controllability analysis as preprocessing method for sensor placement. Based on case studies we demonstrate that the method decreases the number of decision variables for subsequent optimization dramatically to app. 30 to 40 percent.

Operation of remote mobile sensors for security of drinking water distribution systems

The deployment of fixed online water quality sensors in water distribution systems has been recognized as one of the key components of contamination warning systems for securing public health. This study proposes to explore how the inclusion of mobile sensors for inline monitoring of various water quality parameters (e.g., residual chlorine, pH) can enhance water distribution system security. Mobile sensors equipped with sampling, sensing, data acquisition, wireless transmission and power generation systems are being designed, fabricated, and tested, and prototypes are expected to be released in the very near future. This study initiates the development of a theoretical framework for modeling mobile sensor movement in water distribution systems and integrating the sensory data collected from stationary and non-stationary sensor nodes to increase system security. The methodology is applied and demonstrated on two benchmark networks. Performance of different sensor network designs are compared for fixed and combined fixed and mobile sensor networks. Results indicate that complementing online sensor networks with inline monitoring can increase detection likelihood and decrease mean time to detection.