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

Can online particle counters and electrochemical sensors distinguish normal periodic and aperiodic drinking water quality fluctuations from contamination?

Early warning systems monitoring the quality of drinking water need to distinguish between normal quality fluctuations and those caused by contaminants. Thus, to decrease the number of false positive events, normal water quality fluctuations, whether periodic or aperiodic, need to be characterized. For this, we used a novel flow-imaging particle counter, a light-scattering particle counter, and electrochemical sensors to monitor the drinking water quality of a pressure zone in a building complex for 109 days. Data were analyzed to determine the feasibility of the sensors and particle counters to distinguish periodic and aperiodic fluctuations from real-life contaminants. The concentrations of particles smaller than 10 μm and N, Small, Large, and B particles showed sudden changes recurring daily, likely due to the flow rate changes in the building complex. Conversely, the concentrations of larger than 10 μm particles and C particles, in addition to the responses of electrochemical sensors, remained in their low typical values despite flow rate changes. The aperiodic events, likely resulting from an abnormally high flow rate in the water mains due to maintenance, were detected using particle counters and electrochemical sensors. This study provides insights into choosing water quality sensors by showing that machine learning-based particle classes, such as B, C, F, and particles larger than 10 μm are promising in distinguishing contamination from aperiodic and periodic fluctuations while the use of other particle classes and electrochemical sensors may require dynamic baseline to decrease false positive events in an early warning system.

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.

Quality zones automatically identified in water distribution networks by applying data clustering methods to conductivity measurements

This paper presents a clustering study showing how conductivity measured every five minutes by 215 probes over four years can be used to determine specific quality zones for a large Water Distribution Network (WDN): 8500 km of pipes, 4.6 M customers. Conductivity time-series are compared using Dynamic Time Warping. Then, probes are ordered using a density-based method, and probe clusters are extracted automatically. The clusters are a sound representation of water quality in the WDN, both in terms of water origin and water residence time. More specifically, zones directly impacted by plants or by external water imports, mixing zones and zones influenced by tanks, can be isolated and analyzed. Globally, 82% of the probes were found to be clustered, consistent with expert knowledge on the WDN operation; 13% were unclassified; 3% were erroneously clustered; and 1% seemed to be reasonably clustered, without any physical understanding yet. Besides providing users with an increased understanding of water quality in WDNs, conductivity-based clusters offer an interesting prior tool for contamination warning systems.

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.

Prognosis of Water Quality Sensors Using Advanced Data Analytics: Application to the Barcelona Drinking Water Network

Water Utilities (WU) are responsible for supplying water for residential, commercial and industrial use guaranteeing the sanitary and quality standards established by different regulations. To assure the satisfaction of such standards a set of quality sensors that monitor continuously the Water Distribution System (WDS) are used. Unfortunately, those sensors require continuous maintenance in order to guarantee their right and reliable operation. In order to program the maintenance of those sensors taking into account the health state of the sensor, a prognosis system should be deployed. Moreover, before proceeding with the prognosis of the sensors, the data provided with those sensors should be validated using data from other sensors and models. This paper provides an advanced data analytics framework that will allow us to diagnose water quality sensor faults and to detect water quality events. Moreover, a data-driven prognosis module will be able to assess the sensitivity degradation of the chlorine sensors estimating the remaining useful life (RUL), taking into account uncertainty quantification, that allows us to program the maintenance actions based on the state of health of sensors instead on a regular basis. The fault and event detection module is based on a methodology that combines time and spatial models obtained from historical data that are integrated with a discrete-event system and are able to distinguish between a quality event or a sensor fault. The prognosis module analyses the quality sensor time series forecasting the degradation and therefore providing a predictive maintenance plan avoiding unsafe situations in the WDS.

Solving Management Problems in Water Distribution Networks: A Survey of Approaches and Mathematical Models

Modern water distribution networks (WDNs) are complex and difficult to manage due to increased level of urbanization, varying consumer demands, ageing infrastructure, operational costs, and inadequate water resources. The management problems in such complex networks may be classified into short-term, medium-term, and long-term, depending on the duration at which the problems are solved or considered. To address the management problems associated with WDNs, mathematical models facilitate analysis and improvement of the performance of water infrastructure at minimum operational cost, and have been used by researchers, water utility managers, and operators. This paper presents a detailed review of the management problems and essential mathematical models that are used to address these problems at various phases of WDNs. In addition, it also discusses the main approaches to address these management problems to meet customer demands at the required pressure in terms of adequate water quantity and quality. Key challenges that are associated with the management of WDNs are discussed. Also, new directions for future research studies are suggested to enable water utility managers and researchers to improve the performance of water distribution networks.

Real-Time Burst Detection in District Metering Areas in Water Distribution System Based on Patterns of Water Demand with Supervised Learning

This paper proposes a new method to detect bursts in District Metering Areas (DMAs) in water distribution systems. The methodology is divided into three steps. Firstly, Dynamic Time Warping was applied to study the similarity of daily water demand, extract different patterns of water demand, and remove abnormal patterns. In the second stage, according to different water demand patterns, a supervised learning algorithm was adopted for burst detection, which established a leakage identification model for each period of time, respectively, using a sliding time window. Finally, the detection process was performed by calculating the abnormal probability of flow during a certain period by the model and identifying whether a burst occurred according to the set threshold. The method was validated on a case study involving a DMA with engineered pipe-burst events. The results obtained demonstrate that the proposed method can effectively detect bursts, with a low false-alarm rate and high accuracy.

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

Water monitoring technologies are widely used for contaminants detection in wide variety of water ecology applications such as water treatment plant and water distribution system. A tremendous amount of research has been conducted over the past decades to develop robust and efficient techniques of contaminants detection with minimum operating cost and energy. Recent developments in spectroscopic techniques and biosensor approach have improved the detection sensitivities, quantitatively and qualitatively. The availability of in-situ measurements and multiple detection analyses has expanded the water monitoring applications in various advanced techniques including successful establishment in hand-held sensing devices which improves portability in real-time basis for the detection of contaminant, such as microorganisms, pesticides, heavy metal ions, inorganic and organic components. This paper intends to review the developments in water quality monitoring technologies for the detection of biological and chemical contaminants in accordance with instrumental limitations. Particularly, this review focuses on the most recently developed techniques for water contaminant detection applications. Several recommendations and prospective views on the developments in water quality assessments will also be included.

In situ detection of water quality contamination events based on signal complexity analysis using online ultraviolet-visible spectral sensor

The contaminant detection in water distribution systems is essential to protect public health from potentially harmful compounds resulting from accidental spills or intentional releases. As a noninvasive optical technique, ultraviolet-visible (UV-Vis) spectroscopy is investigated for detecting contamination events. However, current methods for event detection exhibit the shortcomings of noise susceptibility. In this paper, a new method that has less sensitivity to noise was proposed to detect water quality contamination events by analyzing the complexity of the UV-Vis spectrum series. The proposed method applied approximate entropy (ApEn) to measure spectrum signals' complexity, which made a distinction between normal and abnormal signals. The impact of noise was attenuated with the help of ApEn's insensitivity to signal disturbance. This method was tested on a real water distribution system data set with various concentration simulation events. Results from the experiment and analysis show that the proposed method has a good performance on noise tolerance and provides a better detection result compared with the autoregressive model and sequential probability ratio test.

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.