Alex Francisque

Evaluating risk of water mains failure using a Bayesian belief network model

It has been reported that since year 2000, there have been an average 700 water main breaks per day only in Canada and the USA costing more than CAD 10 billions/year. Moreover, water main leaks affect other neighboring infrastructure that may lead to catastrophic failures. For this, municipality authorities or stakeholders are more concerned about preventive actions rather reacting to failure events. This paper presents a Bayesian Belief Network (BBN) model to evaluate the risk of failure of metallic water mains using structural integrity, hydraulic capacity, water quality, and consequence factors. BBN is a probabilistic graphical model that represents a set of variables and their probabilistic relationships, which also captures historical information about these dependencies. The proposed model is capable of ranking water mains within distribution network that can identify vulnerable and sensitive pipes to justify proper decision action for maintenance/rehabilitation/replacement (M/R/R). To demonstrate the application of proposed model, water distribution network of City of Kelowna has been studied. Result indicates that almost 9% of the total 259 metallic pipes are at high risk in both summer and winter.

Online Drinking Water Quality Monitoring: Review on Available and Emerging Technologies

Online drinking water quality monitoring technologies have made significant progress for source water surveillance and water treatment plant operation. The use of these technologies in the distribution system has not been favorable due to the high costs associated with installation, maintenance, and calibration of a large distributed array of monitoring sensors. This has led to a search for newer technologies that can be economically deployed on a large scale. This paper includes a brief description of important parameters for drinking water and current available technologies used in the field. The paper also provides a thorough review of the advances in sensor technology for measurement of common water quality parameters (pH, turbidity, free chlorine, dissolved oxygen, and conductivity) in drinking water distribution systems.

Leakage detection and location in water distribution systems using a fuzzy-based methodology

Loss of water due to leakage is a common phenomenon observed practically in all water distribution systems (WDS). However, the leakage volume can be reduced significantly if the occurrence of leakage is detected within minimal time after its occurrence. This paper proposes a novel methodology to detect and diagnose leakage in WDS. In the proposed methodology, a fuzzy-based algorithm has been employed that incorporates various uncertainties into different WDS parameters such as roughness, nodal demands, and water reservoir levels. Monitored pressure in different nodes and flow in different pipes have been used to estimate the degree of membership of leakage and its severity in terms of index of leakage propensity (ILP). Based on the degrees of leakage memberships and the ILPs, the location of the nearest leaky node or leaky pipe has been identified. To demonstrate the effectiveness of the proposed methodology, a small distribution network was investigated which showed very encouraging results. The proposed methodology has a significant potential to help water utility managers to detect and locate leakage in WDS within a minimal time after its occurrence and can help to prioritise leakage management strategies.

Evaluating Water Quality Failure Potential in Water Distribution Systems: A Fuzzy-TOPSIS-OWA-based Methodology

The goal of a water distribution system (WDS) is to deliver safe water with desirable quality, quantity and continuity to the consumers. In some cases, a WDS fails to deliver safe water due to the compromise/ failure of water quality which may have devastating consequences. The frequency and consequence of a water quality failure (WQF) can be reduced if prognostic analysis and necessary remedial measures are taken on time. This study developed a prognostic model to predict WQF potential in a WDS. The study identifies important factors (parameters) which can directly and/or indirectly linked to WQFs. These factors are classified into two groups—the causes of WQF such as lack of free residual chlorine, or excess of total organic carbon, and the symptoms of WQF such as taste & odor, color which are in fact the effects of certain causes of WQF. The interrelationships among the symptoms and the causes have been established based on extensive literature review and elicited expert opinion. A fuzzy-TOPSIS-OWA-based model has been developed to identify the impacts of different influencing parameters on the overall WQF potential. The developed model has been implemented for a WDS in Quebec City (Canada). To study the impacts of uncertainties of the influencing factors, a Monte Carlo simulation-based sensitivity analysis has been carried out. It is anticipated that the developed model can help water utilities to understand the role of different factors on WQF.

Evaluation of source water protection strategies: a fuzzy-based model.

Source water protection (SWP) is an important step in the implementation of a multi-barrier approach that ensures the delivery of safe drinking water. Available decision-making models for SWP primarily use complex mathematical formulations that require large data sets to perform analysis, which limit their use. Moreover, most of them cannot handle interconnection and redundancy among the parameters, or missing information. A fuzzy-based model is proposed in this study to overcome the above limitations. This model can estimate a reduction in the pollutant loads based on selected SWP strategies (e.g., storm water management ponds, vegetated filter strips). The proposed model employs an export coefficient approach and account for the number of animals to estimate the pollutant loads generated by different land usages (e.g., agriculture, forests, highways, livestock, and pasture land). Water quality index is used for the assessment of water quality once these pollutant loads are discharged into the receiving waters. To demonstrate the application of the proposed model, a case study of Page Creek was performed in the Clayburn watershed (British Columbia, Canada). The results show that increasing urban development and poorly managed agricultural areas have the most adverse effects on source water quality. The proposed model can help decision makers to make informed decisions related to the land use and resource allocation.

Evidential reasoning using extended fuzzy Dempster–Shafer theory for handling various facets of information deficiency

This work investigates the problem of combining deficient evidence for the purpose of quality assessment. The main focus of the work is modeling vagueness, ambiguity, and local nonspecificity in information within a unified approach. We introduce an extended fuzzy Dempster–Shafer scheme based on the simultaneous use of fuzzy interval‐grade and interval‐valued belief degree (IGIB). The latter facilitates modeling of uncertainties in terms of local ignorance associated with expert knowledge, whereas the former allows for handling the lack of information on belief degree assignments. Also, generalized fuzzy sets can be readily transformed into the proposed fuzzy IGIB structure. The reasoning for quality assessment is performed by solving nonlinear optimization problems on fuzzy Dempster–Shafer paradigm for the fuzzy IGIB structure. The application of the proposed inference method is investigated by designing a reasoning scheme for water quality monitoring and validated through the experimental data available for different sampling points in a water distribution network.

Online monitoring of drinking water quality in a distribution network: a selection procedure for suitable water quality parameters and sensor devices

Designing a state-of-the-art online water quality monitoring system requires a comprehensive procedure for the selection of pertinent water quality parameters and available cost-effective and reliable sensors. Proper selection of pertinent contaminants/indicators for online monitoring ensures that the overall quality of the distributed water is not compromised and all regulations/guidelines are complied with fully. On the other hand, the selection of monitoring devices and sensors suitable for a typical online water quality monitoring system involves consideration of various factors including reliability, cost, response time, etc. In this paper, we illustrate the use of a multi-criteria decision analysis methodology based on an analytic hierarchy process to optimize the selection of indicator/contaminant sets and related water quality sensors as applied to drinking water distribution networks. A similar methodology could be used for most online monitoring applications in other areas. For the water distribution network application, we shortlisted 18 water quality indicators (WQI) from an initial list of 94 parameters. Similarly, from a comprehensive list of 71 potential water quality monitoring devices, nine were shortlisted for further analysis. Three of these nine monitoring devices were found to be suitable for monitoring 14 of the 18 shortlisted WQI. These three monitoring devices were found to be the best sensors because they demonstrated the ability to monitor 14 selected WQI simultaneously.

Water distribution system failure: a framework for forensic analysis

The main purpose of a water distribution system (WDS) is to deliver safe water of desirable quality, quantity and continuity to consumers. However, in many cases, a WDS fails to fulfill its goal owing to structural and associated hydraulic failures and/or water quality failures. The impact of these failures can be reduced significantly if preventive actions are taken based on their potential of occurrences or if a failure occurs and is detected within a minimum period of time after its occurrence. The aim of this research was to develop a forensic system for WDS failures. As part of the proposed forensic analysis, a framework has been developed, which investigates structural and associated hydraulic failures as well as water quality failures and integrates all failure investigation under a single platform. Under this framework, four different models have been developed to evaluate and identify structural and associated hydraulic failures and water quality failures. If a failure is detected in the system, the framework is capable of identifying the most probable location of the failure. To investigate the effectiveness of the proposed framework, the developed models have been tested and implemented in different WDSs.

Classification of heterotrophic plate counts (HPC) in a water distribution network: a fuzzy rule-based approach

Heterotrophic plate count (HPC) is one of the most common indicators used to monitor microbiological water quality in distribution networks. This paper applies and compares two fuzzy rule-based models to estimate HPC levels in a distribution network using a limited number of water quality parameters. The proposed fuzzy rule-based models include Mamdani and TSK (Takagi, Sugeno, and Kang) algorithms. The models are discussed through a case study of a distribution network (DN) in Quebec City (Canada). Both models properly estimate when HPC levels (datum per datum) are low, however their predictive ability is limited when HPC levels are high. When the outputs (HPC levels) are converted into four classes and the models are used as “classifiers”, their performances are very good. The average percent deviation is lower for the TSK model (15%) than for Mamdani model (17%). Implemented as “classifiers”, both models can be used for identifying vulnerable locations for microbiological contamination within the DN. Given the complexity of the growth of HPC bacteria in water distribution networks and the involvement of numerous contributory factors; results obtained are “promising”. Nevertheless, strategies to improve the models are also discussed.

Networked fuzzy belief rule-based system for spatiotemporal monitoring

This paper introduces a spatiotemporal data aggregation scheme using a novel networked fuzzy belief rule-based (NF-BRB) system. The proposed NF-BRB system is employed to design a decision support tool for relative water quality assessment in the distribution network. Different nodes of the network are grouped in several strata with the connectivity of the nodes shown using a spanning tree. For each stratum, a compact NF-BRB system is designed. The proposed system has flexible and tunable parameters that allow the incorporation of both subjective and numerical information. A learning algorithm to find the locally optimum parameters of the NF-BRB system is employed. The rule aggregation is performed based on a cost-effective approach that uses belief rule utility and disapproval among activated rules. Then, pignistic probabilities of fuzzy evaluation grades are found through aggregation among local strata. The case study of spatiotemporal data aggregation based on hourly data of online monitoring locations of a water distribution network is investigated.