Nurit Oliker

Minimum volume ellipsoid classification model for contamination event detection in water distribution systems

Abstract The presented study features an event detection model alerting for contamination events in water distribution systems. The developed model comprises a minimum volume ellipsoid (MVE) classifier, detecting outlier measurements, and a following sequence analysis utilizing the MVE binary output, for the classification of events. The model is updated continuously and exploits a constantly growing data base. The MVE enables simultaneous analysis of the water quality parameters. The multivariate analysis explores the relations between water quality parameters and detects changes in their common patterns. The suggested model applied an un-supervised classification method, eliminates the need for simulated events examples in the classifier construction. In the absent of satisfying information regarding the influence of contamination event on the parameter measurements, eliminating the use of any assumption contributes to the model reliability and generality. The model was trained on a real water utility data, and tested on randomly simulated events that were superimposed on the original data base. The model showed high accuracy and detection ability compared to previous studies.

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.

Inclusion of Mobile Sensors in Water Distribution System Monitoring Operations

AbstractThis study describes a binary integer programming model for mutually-operating fixed and mobile sensors in water distribution systems. The proposed method applies a deterministic optimization scheme for maximizing the monitored volume within network clusters. For a given budget, the model determines the ratio of mobile sensors to fixed sensors along with their placement and release strategies. Through assessing the benefit of placing each fixed sensor and the time and location of mobile sensors release, the combination of fixed and mobile sensors is determined. Utilizing mobile sensors for water quality monitoring is still in its infancy. Such sensors are equipped with self-powered sensing, sampling, data acquisition, and wireless transmission units. The model initiates with the combined operation of mobile and fixed sensors. It then explores the benefits of mobile sensors compared to fixed. The two battle of the water sensor networks (BWSN) are utilized for demonstrating the model’s capabilities....

Spatial event classification using simulated water quality data

This study deals with the integration of contamination simulations and a spatial event detection model. The simulation of contaminant intrusion includes detailed chemical-specific reactions within a multi-species water quality model. This set-up generates a scenario of contaminant distribution and produces a continuous multiple sensor stations database. Three organophosphates pesticides, Chlorpyrifos, Malathion, and Parathion, are modeled as possible contaminants. The event detection model comprises both local and spatial data analysis. The local model applies a previously developed single-sensor event detection model with a higher alert threshold that reduces false alarm rates. The spatial model considers upstream sensor datasets which are examined for their uniqueness and mutual resemblance in a sliding time window. The model utilizes outlier detection, data analyses, and network hydraulics for the detection of suspicious spatial trends. The proposed algorithm is capable of detecting events with low contamination signatures and spatial influence. Two case studies are explored and compared to the single sensor model. The proposed methodology resulted in a lower number of false alarms compared to the previous single sensor event detection modeling approach. Display Omitted Integration of event detection model with simulated contamination events.Spatial event detection and classification.Sensor measurements generated by chemistry simulations of organophosphates.

Water distribution systems event detection

Since the events of 9/11 2001 in the US the world public awareness to possible terrorist attacks on water supply systems has increased dramatically, causing the security of drinking water distribution systems to become a major concern around the globe. Among the different threats, a deliberate chemical or biological contaminant injection is the most difficult to address, both as a consequence of the uncertainty surrounding the type of the injected contaminant and its consequences, as well as the uncertainty of location and time of the injection. In principle, a pollutant can be injected at any water distribution system connection (node) using a pump or a mobile pressurized tank. Although backflow preventers provide an obstacle to such actions, they do not exist at all connections, and at some might not be functional. This paper describes recent effort modeling of Avi Ostfelds research team on water distribution systems event detection. The basic event detection framework is entitled AEDA (Aquatic Event Detection Algorithm) which utilizes Artificial Neural Networks (ANNs) for studying the interactions between multivariate water quality parameters and detecting possible outliers. Other layers on top of AEDA explore tradeoffs among contamination event parameters and improving its performance capabilities. Those and AEDA are reviewed in this paper.

Mobile Sensors for Water Quality Management in Water Distribution Systems

Delivery of safe drinking water to consumers is a vital infrastructure of all populations. Any large water distribution system is inherently prone to fault from intentional or accidental contamination, placing large populations at risk. Monitoring these systems through a wireless network of stationary sensors (WSN) has shown to be an effective method to protect a water supply. Recent technological advancements allow for the implementation of a high resolution mobile wireless sensor network (MWSN); where sensors function within the water flowing through municipal pipes to measure water quality parameters and to transmit data to fixed ground transceivers. With mobile sensor prototypes being developed and tested, a MWSN is likely to be physically deployed in the near future. Previous work has shown an ideal MWSN to increase water security system performance. Accounting for uncertainties in: data collection, data transmission to fixed transceivers and sensor lifetimes will provide beneficial insight to the realistic performance of a MWSN. The non-ideal operation of mobile sensors is simulated and applied to sample municipal networks using EPANET and genetic algorithms to optimize the deployment of multiple mobile sensors, and quantify operational sensitivity. Results show a MWSN used for protection of public water supply to be highly sensitive to battery life and receiver network coverage, while the interval between measurements shows little affect to MWSN performance.

A frequency based transit assignment model that considers online information at the boarding stop

This paper develops a frequency based transit assignment model, assuming that transit arrival times are available to the passenger at the boarding stop. The model considers the estimated arrival times together with the expected travel times, when selecting a path. The assignment procedure includes the setting of decision rules for different cases of arrival times, and the calculation of probabilities for these different cases. The model is applied on a real-size network, and the results are compared to the well-known optimal strategies method. The suggested model showed significantly different results and a notable reduction in the total travel time. The results illustrate the potential impact of online information on travel behavior, and emphasize the need of its consideration in assignment models.

Water Distribution Systems Complex Contamination Simulations for Event Detection Model Calibration and Verification

Event detection is currently one of the most challenging topics in water distribution systems analysis. The problem is related to how regular on-line hydraulic (e.g., pressure, flow) and water quality (e.g., pH, free chlorine, conductivity, oxidation reduction potential, temperature) measurements can be efficiently utilized to designate a contamination occurrence. To date all experiments on testing event detection models are based on generating arbitrary manipulations on water quality parameters (e.g., through utilization of Gaussian probability density functions). Employment of such an approach is on the one hand generic, but on the other has no physical connection to the water distribution system physical behaviour. In fact, all event detection models to date for water distribution systems do not take advantage of the hydraulic/water quality understanding of the system in the decision process of event detection. This study describes some initial steps of utilizing EPANET-MSX for generating test contamination scenarios aimed at more realistically validate event detection models through complex contamination events simulation modelling. INTRODUCTION This work is aimed at implementing simulated contamination events in water distribution systems for exploring their impact on water quality parameters. The simulation outcomes are planned to be linked to an event detection model for calibrating its parameters and explore its performance capabilities. Studies that were conducted in the field so far (e.g., Perelman et al., 2012; CANARY@software.sandia.gov/trac/canary) have dealt with the development of contamination event detection models, by the promise that contamination events cause some unknown changes in the water quality parameters. Respectively, the events simulated for training and testing of the models were composed of artificial random disturbances that were simulated and superimposed upon the original data. In this study it is suggested that the simulations modeling the hydraulic and water quality behavior within the water distribution systems under conditions of various contaminants intrusion of complex compounds will be generated using a physical model. Chemical reactions, as well as free chlorine loss, formation of disinfection byproducts and residuals, species dynamics, and adsorption on pipe walls were taken 1016 World Environmental and Water Resources Congress 2013: Showcasing the Future