Dissertations / Theses on the topic 'Water distribution networks maintenance'

Submitted by Caroline Falcao (caroline.rfalcao@ufpe.br) on 2017-06-01T17:12:43Z No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Marcele Elisa.pdf: 2410134 bytes, checksum: fb1c34a2e156d50ef7918acf93596973 (MD5)
Made available in DSpace on 2017-06-01T17:12:43Z (GMT). No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Marcele Elisa.pdf: 2410134 bytes, checksum: fb1c34a2e156d50ef7918acf93596973 (MD5) Previous issue date: 2012-06-13
Devido ao envelhecimento e depreciação dos componentes de uma rede de distribuição de água é comum ocorrem vazamentos e rupturas na rede, ocasionando perdas e podendo acarretar a contaminação da água e descontinuidade no seu fornecimento pela interrupção do abastecimento de água durante a reparação da infraestrutura, bem como a redução da água disponível no sistema, gerando a insatisfação dos seus usuários. Nestes casos, a setorização da rede torna mais fácil qualquer atividade de manutenção, além de atingir um menor número de unidades de consumidoras pela interrupção do abastecimento. Porém, além do elevado custo da alocação de válvulas de fechamento, as Normas Brasileiras sobre setorização das redes dizem respeito apenas a questões técnicas que devem ser obedecidas pelos setores de manobra. Portanto, um modelo de setorização de manobra que leva em consideração as características das unidades consumidoras é proposto. Com ele pretende-se uma setorização que respeite as restrições previstas por lei, busque a maximização dos benefícios gerados pela setorização às unidades consumidoras e a minimização do número de válvulas de fechamento necessárias, ou seja, dos custos. Para medir o impacto da falta de água é gerado um índice de priorização que agrega as características das unidades consumidoras por meio de um método de agregação multicritério, o SMARTER. As simulações realizadas mostram que apenas a alocação inicial das válvulas pela NBR 12218/94 não garante a maximização dos benefícios gerados pela setorização da rede aos usuários. Neste caso, com o modelo proposto pode-se encontrar soluções melhores.
Due to aging and depreciation of the water distribution network components, leaks and ruptures are common on the water network. These cause the water loss, sometimes the water contamination and the discontinuity in water supply, due the supply interruption during the infrastructure repair, as well as reduction of available water in the system, causing the users' dissatisfaction. In these cases, network segmentation makes any maintenance activity easy and this achieves a smaller number of consuming units by water supply interruption. However, besides the high valves allocation cost, the Brazilian norms about network segmentation relate only technical issues that must be followed. Therefore, a network segmentation model, which takes into account the characteristics of consumer units, is proposed. It aims a network segmentation which complied the restrictions provided by law, it seeks the maximum benefit to consumer units, generated by the segmentation, and it seeks minimum number of valves required, i.e., the costs. To measure the impact on water lack is generated a priority index which adds the characteristics of consuming units via a multicriteria aggregation method, the SMARTER. The simulations realized showed that only the initial valves allocation, by the NBR 12218/94, does not guarantee the maximization of the benefits generated by the network segmentation for users. In this case, with the proposed model, better solutions can be found.

Water distribution network models are used for different purposes. In this study, a model, used for daily operational issues is concerned. Models results should be consistent with actual conditions for sound decisions during operational studies. Adjusting model parameters according to site measurements in order to fit the model to obtain realistic results is known as calibration. Researchers have carried out numerous studies on calibration and developed various methods. In this study, an actual network (N8.3 Pressure Zone, Ankara) has been calibrated by two classical methods developed by Walski (1983) and Bhave (1988). The network parameter calibrated in this study is Hazen-Williams roughness coefficient, C-factor, and other parameters have been lumped in the C-factor. Results of the analysis showed that, C-factors have been found in a wide range.

With continuously increasing urbanization, consumer demands and expansion of water supply systems, determination of efficient pump schedules became a more difficult task. Pumping energy costs constitute a significant part of the operational cost of the water distribution networks. This study aims to provide an effective daily pump schedule by minimizing the energy costs for constant and also for multi tariff of electricity (3 Kademeli Elektrik Tarifesi) in water distribution network. A case study has been performed in an area covering N8.3 and N7 pressure zones which are parts of Ankara water distribution network. Both pressure zones consists of 3 multiple pumps in pump station and one tank having 5000 m3 storage volume each. By using genetic algorithm based software (WaterCAD Darwin Scheduler) least-cost pump scheduling and operation policy for each pump station has been determined while satisfying target hydraulic performance requirements such as minimum and maximum service pressures, final water level of storage tank and maximum velocity in pipeline. 32 different alternative scenarios have been created which include multi tariff energy prices, constant tariff energy price, insulated system condition, uninsulated system condition and different pump combinations. The existing base scenario and alternative scenarios which were prepared by using optimal pump schedules have been compared and the achievements of optimizing pump operation have been analyzed. At the end of the study, a satisfying result has been observed that by using determined optimal pump schedule, minimum % 14 of total energy cost can be saved in existing water supply system.

Resilience is widely interpreted as the capacity of a system to resist (preparation phase), absorb and withstand (responding phase), and rapidly recover from (restoration phase) exceptional conditions. During this study, a mixed variety of calculations were assessed in order to find the best solution for determination of resilience and reliability of a simple network. Then, Todini’s formula and failure index was applied to estimate reliability of system in different scenarios as constant demands in period of 24 hours, constant demands in period of 72 hours when tank will be empty (failure of tank), variable demands in period of 24 hours. At first hydraulic simulation of those scenarios was done by EPANET and validated by MATLAB-TOOLKIT. Then, Resilience index (RI), Failure Index (FI) and reliability (R) of system were measured. Finally, an optimization procedure was done to make a water distribution network with highest resilience and lowest failure probability. All these procedures have been applied on a real network as WDS of Modena. It concluded that this method can be used for every water system without considering the type of failure. As a result, first scenario has a constant decreased and increased trend of RI and FI respectively because of diminishing of water level in tank. During second scenario, there is significant change after the tank will be empty (or it is broke). Third scenario is more like a real network with variable demand during a day. It was concluded that there is a minimum resilience parameter during day when a peak time of water demand expected. It can be justified that the reservoir and pump system had to sustain more pressure to satisfy the demands of junctions. Maximum resilience is related to night during a day with less demands and providing water by tank to other junctions. This modelling could be useful to optimize the dimensions and features of instruments to increase availability and reliability of system.

Disinfection of the municipal water systems is mostly achieved by means of chlorine addition at water treatment plants known as sources. Thus, there should be an adequate chlorine concentration at the source for an effective disinfection throughout the system by considering upper and lower limits of disinfectant. However, since the disinfectants are reactive and decays through the system, chlorine added at the source may not be enough to maintain desired disinfectant residuals which may lead to water quality problems in the water distribution system. Moreover, the disinfectants such as chlorine has also an effect to be carcinogen due to formation of disinfectant by-products. Thus, the system should balance the amount of disinfectant supplied while minimizing the health risk. In such a case, it is recommended that one or more booster disinfection stations can be located throughout the system. Such a method can provide more uniform distribution of the chlorine concentration while reducing the amount of the disinfectant used. In this thesis, optimum scheduling, and injection rates of the booster disinfection stations have been searched. The objective is to minimize the injected mass dosage rate subjected to the provision of adequate and more uniform residual concentration in the network. Determination of variable network hydraulics and chlorine concentrations is held out by EPANET network simulation sofware. A C++ code was developed to interface with EPANET by means of the EPANET Programmer'
s Toolkit for linear optimization of the disinfectant mass dosage rate applied to the network.

Water distribution network models are widely used for various purposes such as long-range planning, design, operation and water quality management. Before these models are used for a specific study, they should be calibrated by adjusting model parameters such as pipe roughness values and nodal demands so that models can yield compatible results with site observations (basically, pressure readings). Many methods have been developed to calibrate water distribution networks. In this study, Darwin Calibrator, a computer software that uses genetic algorithm, is used to calibrate N8.3 pressure zone model of Ankara water distribution network
in this case study the network is calibrated on the basis of roughness parameter, Hazen Williams coefficient for the sake of simplicity. It is understood that there are various parameters that contribute to the uncertainties in water distribution network modelling and the calibration process. Besides, computer software&rsquo
s are valuable tools to solve water distribution network problems and to calibrate network models in an accurate and fast way using automated calibration technique. Furthermore, there are many important aspects that should be considered during automated calibration such as pipe roughness grouping. In this study, influence of flow velocity on pipe roughness grouping is examined. Roughness coefficients of pipes have been estimated in the range of 70-140.

The subdivision of water distribution networks (WDN) into zones, known as district metered areas (DMAs), is a popular approach to leakage management used by water companies. The DMAs are formed by permanently closing isolation valves at the boundaries of each zone (known as boundary valves). By forming discrete zones in the WDN, leakage estimates can be made at night when demand is low, which is used to prioritise pipe repair and replacement programmes. However, the permanent closure of boundary valves has also caused several disadvantages, including reduced network resilience to failure, sub-optimal pressure management, and water quality problems. This thesis introduces a novel approach to the operational management of WDNs, where DMAs are dynamically aggregated for improved network resilience, pressure management and water quality, and segregated for leakage monitoring at night. This is facilitated by replacing closed boundary valves with self-powered, remote control valves (dynamic boundary valves). The operation of a dynamic topology can therefore successfully eliminate the disadvantages of conventional DMAs, whilst retaining or improving their success in leakage monitoring. The investigation is carried out both analytically and experimentally on a real, operational UK network supplying approximately 8,000 properties in order to establish the benefits and challenges of the proposed approach. An extensive review of the current and emerging forms of pressure and DMA management from around the world is carried out. A case study using a real large scale network (3,148km of pipeline) demonstrates the current state of DMAs and their conformity to DMA design guidelines, and identifies how a dynamic DMA topology can improve network performance. A novel resilience index (the Reserve Capacity) is then used in the design of the dynamic DMA topology in the experimental programme. The analytical and experimental investigation has demonstrated a 27% reduction in leakage using a dynamic DMA topology over the most common approach to pressure management, and strong improvements in network resilience to failure where 1,400 customers maintained a supply during a real, major burst incident that would otherwise have been disconnected. In order to actuate near optimal control in the experimental programme, a novel optimisation algorithm based on sequential convex programming (SCP) is proposed for the control of valves. The SCP method takes advantage of computationally efficient solvers that facilitate prompt and reliable convergence. The algorithm also includes the development of a novel, convex valve model that can be integrated into efficient null space algorithms. In order to actuate the control, a novel approach for valve control in DMAs with dynamic topology is proposed, where time varying flow modulation curves are used based on the dynamic connectivity of DMAs.

Water distribution networks are a fundamental part of any modern city and their daily operations constitute a significant expenditure in terms of energy and maintenance costs. Careful scheduling of pump operations may lead to significant energy savings and prevent wear and tear. By means of computer simulation, an optimal schedule of pumps can be found by an optimisation algorithm. The subject of this thesis is the study of pump scheduling as an optimisation problem. New representations of pump schedules are investigated for restricting the number of potential schedules. Recombination and mutation operators are proposed, in order to use the new representations in evolutionary algorithms. These new representations are empirically compared to traditional representations using different network instances, one of them being a large and complex network from UK. By means of the new representations, the evolutionary algorithm developed during this thesis finds new best-known solutions for both networks. Pump scheduling as the multi-objective problem of minimising energy and maintenance costs in terms of Pareto optimality is also investigated in this thesis. Two alternative surrogate measures of maintenance cost are considered: the minimisation of the number of pump switches and the maximisation of the shortest idle time. A single run of the multi-objective evolutionary algorithm obtains pump schedules with lower electrical cost and lower number of pump switches than those found in the literature. Alternatively, schedules with very long idle times may be found with slightly higher electrical cost. Finally, ant colony optimisation is also adapted to the pump scheduling problem. Both Ant System and Max-Min Ant System are tested. Max-Min Ant System, in particular, outperforms all other algorithms in the large real-world network instance and obtains competitive results in the smallest test network. Computation time is further reduced by parallel simulation of pump schedules.

The thesis is a treatise of the quantity and quality aspects of potable water in distribution systems. The privatisation of the UK Water Industry in 1989 has seen the requirement for the Water Companies in England and Wales to be responsible for the delivery of good quality water that meets the demand of all consumers. In respect of the quantity of supply, there have been many previous studies that have examined the hydraulic performance of distribution systems and there are now many proprietary mathematical models that have been successfully used in this study. However, in respect of water quality the literature review has highlighted that the modelling approach is not so well advanced, as water quality is a function of many concepts, processes and parameters that include the source and age of water, the condition and deterioration of the assets in the system, the microbiological, chemical and physical processes and the network hydraulic performance, including pressure transients. These processes are highly interactive and complex. In an attempt to better understand these processes a programme of research has been completed that has involved a field evaluation of the performance of a live system, including the development of instrumentation to continually measure water quality, and the development of a mathematical model to describe the processes associated with the age of water and the propagation of conservative and non-conservative substances. An initial attempt has also been made to develop a micro-biological model and a sediment transport model. New original concepts developed by the author include age, biological and diagnostic models that may be used to identify the source of any incident (hydraulic or pollution) and the application of the model in near real time.

Central to the South African government’s vision of providing services to all is on-going maintenance of public infrastructure. Since 1994 the government focused on addressing backlogs in the provision of water services through new infrastructure investment; however it failed to make sufficient investment in the maintenance and renewal of this infrastructure (SAICE, 2006). Older infrastructure is not being renewed or refurbished as required and planned preventative maintenance on new infrastructure is inadequate (SAICE, 2006). This has been generally attributed to poor management strategies that are exacerbated by lack of skills in water services utilities and low levels of funding provisions (Mescht & Jaarsveld, 2012; FFC, 2013). The continuing poor maintenance of water distribution networks has contributed to high leakage rates in South Africa (FFC, 2013; DBSA, 2012). To address challenges of maintenance of water distribution infrastructure a regulatory framework to guide municipalities is critical. The government approved the National Infrastructure Management Strategy (NIMS) in 2006 to support simultaneous infrastructure investment and maintenance (CIDB, 2008). One of the key thrusts of the strategy is the strengthening of the regulatory framework that governs planning and budgeting for maintenance. The literature survey of this study found that initiatives associated with the NIMS were very slow in gaining traction. The study reviews water services infrastructure management frameworks that are based on present legislative instruments and standards for two study areas; City of Capetown and City of Johannesburg. Challenges associated with effective management of water distribution infrastructure are assessed based on established infrastructure management policies, strategies and asset management plans for each entity. For each study area leakage control strategies are the key maintenance strategy outputs associated with the implementation of the management frameworks; therefore the study reviewed sector plans and annual reports to assess challenges associated with carrying out effective maintenance. The findings of the study show a correlation between the adoption of maintenance management strategies and the improvement of the performance of water distribution networks for both Cape Town and Johannesburg. The strategies are driven at the highest level of decision making in the municipalities as budgeting requirements are supported by the Integrated Development Plans of each study area. The maintenance allocations however remain below the international benchmark to enable the municipal entities to carry out satisfactory maintenance of their distribution infrastructure.

This study gives a description about the development of a computer model, RealPipe, which relates genetic algorithm (GA) to the well known problem of least-cost design of water distribution network. GA methodology is an evolutionary process, basically imitating evolution process of nature. GA is essentially an efficient search method basically for nonlinear optimization cases. The genetic operations take place within the population of chromosomes. By means of various operators, the genetic knowledge in chromosomes change continuously and the success of the population progressively increases as a result of these operations. GA optimization is also well suited for optimization of water distribution systems, especially large and complex systems. The primary objective of this study is optimization of a water distribution network by GA. GA operations are realized on a special program developed by the author called RealPipe. RealPipe optimizes given water network distribution systems by considering capital cost of pipes only. Five operators are involved in the program algorithm. These operators are generation, selection, elitism, crossover and mutation. Optimum population size is found to be between 30-70 depending on the size of the network (i.e. pipe number) and number of commercially available pipe size. Elitism rate should be around 10 percent. Mutation rate should be selected around 1-5 percent depending again on the size of the network. Multipoint crossover and higher rates are advisable. Also pressure penalty parameters are found to be much important than velocity parameters. Below pressure penalty parameter is the most important one and should be roughly 100 times higher than the other. Two known networks of the literature are examined using RealPipe and expected results are achieved. N8.3 network which is located in the northern side of Ankara is the case study. Total cost achieved by RealPipe is 16.74 percent lower than the cost of the existing network
it should be noted that the solution provided by RealPipe is hydraulically improved.

Under the envisioned smart city paradigm, there is an increasing demand for the coordinated operation of our infrastructure networks. In this context, this thesis puts forth a comprehensive toolbox for the optimization of electric power and water distribution networks. On the analytical front, the toolbox consists of novel mixed-integer (non)-linear program (MINLP) formulations; convex relaxations with optimality guarantees; and the powerful technique of McCormick linearization. On the application side, the developed tools support the operation of each of the infrastructure networks independently, but also towards their joint operation. Starting with water distribution networks, the main difficulty in solving any (optimal-) water flow problem stems from a piecewise quadratic pressure drop law. To efficiently handle these constraints, we have first formulated a novel MINLP, and then proposed a relaxation of the pressure drop constraints to yield a mixed-integer second-order cone program. Further, a novel penalty term is appended to the cost that guarantees optimality and exactness under pre-defined network conditions. This contribution can be used to solve the WF problem; the OWF task of minimizing the pumping cost satisfying operational constraints; and the task of scheduling the operation of tanks to maximize the water service time in an area experiencing electric power outage. Regarding electric power systems, a novel MILP formulation for distribution restoration using binary indicator vectors on graph properties alongside exact McCormick linearization is proposed. This can be used to minimize the restoration time of an electric system under critical operational constraints, and to enable a coordinated response with the water utilities during outages.
Master of Science
The advent of smart cities has promoted research towards interdependent operation of utilities such as water and power systems. While power system analysis is significantly developed due to decades of focused research, water networks have been relying on relatively less sophisticated tools. In this context, this thesis develops Advanced efficient computational tools for the analysis and optimization for water distribution networks. Given the consumer demands, an optimal water flow (OWF) problem for minimizing the pump operation cost is formulated. Developing a rigorous analytical framework, the proposed formulation provides significant computational improvements without compromising on the accuracy. Explicit network conditions are provided that guarantee the optimality and feasibility of the obtained OWF solution. The developed formulation is next used to solve two practical problems: the water flow problem, that solves the complex physical equations yielding nodal pressures and pipeline flows given the demands/injections; and an OWF problem that finds the best operational strategy for water utilities during power outages. The latter helps the water utility to maximize their service time during power outages, and helps power utilities better plan their restoration strategy. While the increased instrumentation and automation has enabled power utilities to better manage restoration during outages, finding an optimal strategy remains a difficult problem. The operational and coordination requirements for the upcoming distributed resources and microgrids further complicate the problem. This thesis develops a computationally fast and reasonably accurate power distribution restoration scheme enabling optimal coordination of different generators with optimal islanding. Numerical tests are conducted on benchmark water and power networks to corroborate the claims of the developed formulations.

In recent years, water utilities have placed a greater emphasis on the reliability and resilience of their water distribution networks. This focus has increased due to the continuing aging of such infrastructure and the potential threat of natural or man-made disruptions. As a result, water utilities continue to look for ways to evaluate the resiliency of their systems with a goal of identifying critical elements that need to be reinforced or replaced. The simulation of pipe breaks in water reliability studies is traditionally modeled as the loss of a single pipe element. This assumes that each pipe has an isolation valve on both ends of the pipe that can be readily located and operated under emergency conditions. This is seldom the case. The proposed methodology takes into account that multiple pipes may be impacted during a single failure as a result of the necessity to close multiple isolation valves in order to isolate the “segment” of pipes necessary to contain the leak. This document presents a simple graphical metric for use in evaluating the performance of a system in response to a pipe failure. The metrics are applied to three different water distribution systems in an attempt to illustrate the fact that different pipe segments may impact system performance in different ways. This information is critical for use by system managers in deciding which segments to prioritize for upgrades or replacement.

The overall aim of this research has been to develop new algorithms and computer software that may be used to assess the reliability of water distribution systems. Such a tool can be used by design engineers to create systems which are both economical in total cost commensurate with meeting targets for a specified level of reliability. The introduction describes how water supply and distribution systems are normally designed, what they comprise and problems associated with failure or lack of availability of an adequate supply to the end user. This is followed by a resume of current methods and algorithms for the analysis of networks and a detailed examination of the previous work on network optimisation and reliability. Three main algorithms exist for the analysis of water networks. These are the Hardy-Cross methods, the Newton-Raphson methods and the Linear method. A computer program based on the Linear method, which is known to be the most reliable, is proposed for the hydraulic analysis part of the present work. With respect to reliability, a full discussion of the topic, including all the various factors which influence it such as the stochastic nature of customer demands, the apparently random occurrence of pipe breakages and the concept of repair time, is presented. A reliability analysis model, that incorporates simultaneously the three reliability factors mentioned, for the assessment of nodal and system availabilities, is proposed, from which an efficient computer program has been developed and tested. Two models for the design of optimal water distribution systems, based on reliability criteria, have been developed, programmed and tested. The first model makes use of the entropy principle for producing 'reliable' distributions of flow and the Linear Programming technique is used for computation of the least cost design. In the second model, however, a Genetic Algorithm procedure, that incorporates the new reliability analysis model and which is superior to other models has been formulated. The thesis concludes with a comparison between the two methods formulated as a result of this research and applied to realistic practical systems, plus suggestions for further work to improve the optimisation of water distribution networks.

Several experimental and modelling studies have established that leak areas are mostly not fixed but vary linearly with pressure. Introducing this linear relationship into the orifice equation, results in a two-part modified orifice equation for leakage modelling with pressure head exponents of 0.5 and 1.5 respectively. Current hydraulic network solvers apply the conventional power leakage equation to model pressure dependent demands such as leakage. The empirically derived power leakage equation does not explicitly consider the leak area variation with pressure and it has been found to be flawed under certain conditions. The aim of this study therefore, was to incorporate the modified orifice equation into the algorithm of a hydraulic network solver and evaluate the impact this has on leakage modelling. Epanet is the hydraulic modelling software whose algorithm of the network solver was modified. In addition, a stochastic model for network leak generation and distribution was developed. The conventional and the modified software were applied to different levels of stochastically generated and distributed leakage in three differently sized pipe networks. It was found that the conventional power leakage equation results in significant leakage volume and flow rate errors under certain conditions. A methodology was also developed to correct the conventional power leakage equation so that it can be used to model leakage realistically without a change of the software to one that uses the modified orifice equation. The methodology was thereafter applied to an existing model that detects leaks in standard water distribution pipe networks, and the results showed significant improvements in the performance of the model.

The occurrence of discoloured drinking water at customers’ taps, which is mainly caused by the deposition and release of iron (Fe) and manganese (Mn) in water distribution networks (WDNs), is a major concern for both customers and water companies. Increased concentrations of Fe and Mn in WDNs can lead to penalisation by the Drinking Water Inspectorate (DWI) and Water Services Regulation Authority in England and Wales (Ofwat). These high concentration levels can cause aesthetic problems such as giving water an unpleasant metallic taste and staining of laundry. It has also been found that increased Mn concentrations in drinking water can reduce intellectual function of children. Despite efforts by water companies to comply with standards for drinking water, they continue to receive customer complaints related to water discolouration. Currently, most water companies identify high-discolouration-risk regions in WDNs by either selecting areas in the network with high concentrations of Fe and Mn from their routine sampling, or using data obtained from customer complaints related to discolouration. However, these risk assessment methods are imprecise, because only few selected nodes are sampled and not all customers who experience water discolouration complain. Moreover, considering that the water mains in England and Wales span approximately 315,000 km, monitoring Fe and Mn concentrations will always be a difficult and expensive task. It is therefore imperative for water companies to gain a practical understanding of the processes and mechanisms that lead to water discolouration, and to develop a model to identify the high-risk areas in WDNs so that remedial measures can be effectively implemented. The factors that influence Fe and Mn accumulation from post-treatment to customers’ taps through WDNs can be categorised into physical, chemical and biological. However, to date, researchers have only studied these factors partially or separately, but never in combination. None of the current models are able to predict discolouration/Fe and Mn accumulation potential for every node in WSZs using chemical, biological, and hydraulic/physical variables. This study took a holistic approach in investigating these factors. A five-year data set comprising of 36 water quality, hydraulic, and pipe-related variables covering 176 different district metered areas (DMAs) were analysed to identify relevant variables that influence Fe and Mn accumulation potential. Customer complaint data were also investigated for seasonal trends. Majority of the DMAs (67.44%) showed significant peaks in customer complaints during summer. These spikes may be attributed to increased water consumption and warmer water temperatures during this period. An artificial neural network (ANN) model was developed using relevant variables identified through the data analysis. The model could predict Fe and Mn accumulation potential values for every node in a given water supply zone (WSZ). From the risk maps generated by the ANN model, it was observed that most of the regions in the network with high Fe and Mn accumulation potential also had high levels of customer complaints related to discolouration. Although the ANN model could predict Fe and Mn accumulation potential failures in WSZs, its black-box nature made it difficult to explain the causes of the failures, unless they were manually investigated. To overcome the limitation in the ANN model, a fuzzy inference system (FIS) was developed to predict Fe and Mn accumulation potential for every node in WDNs and also capture the chemical, biological and physical processes as water travels through the network. The rules and weights of the rules for the FIS were calibrated using a genetic algorithm. The FIS is also able to determine the causes of the Fe and Mn accumulation potential failures. The ability of the developed models in this research to predict and indicate the causes of high Fe and Mn accumulation potential at the node level make them a unique and practical tool to detect high risk nodes in all regions in WDNs, including regions which have not been sampled. Both models could be of great benefit to water resource engineers and drinking water supply companies in managing water discolouration. They could also be used to investigate variables that influence physical, chemical and biological processes in WDNs.

In this dissertation, we consider two specific types of problems over networks. In the first problem, we explore systematic methods to address some of the challenges in largescale maintenance planning in integrated chemical sites. In the second problem, we investigate different optimization model formulations for the design of flow distribution networks where the flow is potential-driven and nonlinearly related to the potential loss. Maintenance turnaround in the processing industry is a complex asset renewal project that includes huge capital expenditures and downtime losses. The option of deferring or rescheduling a turnaround project typically provides immediate financial relief from capital expenditure. However, the risk of running into site-wide disruptions in the form of unplanned events, yield, and reliability losses is not straightforward to assess. We propose mathematical optimization models to evaluate the risk of loss from turnaround deferrals in integrated sites and provide alternatives to reliably operate the site in a medium-term horizon. In the first chapter, we introduce the turnaround planning problem and the challenges it poses in integrated sites. We also introduce the background for the network design problem. In the second chapter, we study the financial impact of rescheduling turnarounds and the associated risk under unplanned outages. We compare the risk profiles presented by different production planning strategies. We propose a stochastic programming model for production planning that optimally builds up inventory ahead of time to hedge against production losses during unplanned outages. In the third chapter, we extend the stochastic optimization to handle a large set of scenarios and propose a Lagrangean decomposition method that improves a myopic production plan. The fourth chapter proposes a mixed-integer linear programming model that prescribes turnaround schedules when the underlying assets undergo yield loses and selectivity degradation. Here, we also study the impact of deferrals over a long-term horizon. The penultimate chapter addresses the nonlinear network design problem.The closing chapter summarizes the work and provides a few future directions. In the spirit of advancing manufacturing paradigms, the thesis supports investment in modeling efforts that address enterprise-wide planning problems.

Thesis (PhD) - Faculty of Life and Social Sciences, Swinburne University of Technology, 2009.
Submitted in full requirement for the degree of Doctor of Philosophy, Faculty of Life and Social Sciences, Swinburne University of Technology - 2009. Typescript. Includes bibliographical references (p. 137-160)

Two of the most important environmental challenges in the 21st century are to protect the quality of fresh water resources and to utilize renewable energy sources to lower greenhouse gas emissions. This study contributes to the solution of the first challenge by providing methodologies for optimal design of real-time water quality monitoring systems and interpretation of data supplied by the monitoring system to identify potential pollution sources in river networks. In this study, the optimal river water quality monitoring network design aspect of the overall monitoring program is addressed by a novel methodology for the analysis of this problem. In this analysis, the locations of sampling sites are determined such that the contaminant detection time is minimized for the river network while achieving maximum reliability for the monitoring system performance. The data collected from these monitoring stations can be used to identify contamination source locations. This study suggests a methodology that utilizes a classification routine which associates the observations on a contaminant spill with one or more of the candidate spill locations in the river network. This approach consists of a training step followed by a sequential elimination of the candidate spill locations which lead to the identification of potential spill locations. In order to contribute the solution of the second environmental challenge, this study suggests utilizing available excess energy in water distribution systems by providing a methodology for optimal design of energy recovery systems. The energy recovery in water distribution systems is possible by using micro hydroelectric turbines to harvest available excess energy inevitably produced to satisfy consumer demands and to maintain adequate pressures. In this study, an optimization approach for the design of energy recovery systems in water distribution networks is proposed. This methodology is based on finding the best locations for micro hydroelectric plants in the network to recover the excess energy. Due to the unsteady nature of flow in water distribution networks, the proposed methodology also determines optimum operation schedules for the micro turbines.

The development of wireless sensor networks (WSNs) is making it possible to monitor our cities. Due to the small size of the sensor nodes, and their capabilities of transmitting data remotely, they can be deployed at locations that are not easy or impossible to access, such as the pipelines of water distribution networks (WDNs), which plays an important role in protecting environment and securing public health.   The design of WSNs for WDNs faces major challenges. Generally, WSNs are resource-limited because most of the sensor nodes are battery powered. Thus, their resource allocation has to be carefully controlled. The thesis considers two prominent problems that occur when designing WSNs for WDNs: scheduling the sensing of the nodes of static WSNs, and sensor placement for mobile WSNs. These studies are reported in the thesis from three published or submitted papers. In the first paper, the scheduling of sleep/sensing for each sensor node is considered to maximize the whole WSNs lifetime while guaranteeing a monitoring performance constraint. The problem is transformed into an energy balancing problem, and solved by a dynamic programming based algorithm. It is proved that this algorithm finds one of the optimal solutions for the energy balancing problem. In the second paper, the question of how the energy balancing problem approximates the original scheduling problem is addressed. It is shown that even though these two problems are not equivalent, the gap of them is small enough. Thus, the proposed algorithm for the energy balancing problem can find a good approximation solution for the original scheduling problem. The second part of the thesis considers the use of mobile sensor nodes. Here, the limited resource is the number of available such mobile nodes. To maximize the monitoring coverage in terms of population, an optimization problem for determining the releasing locations for the mobile sensor nodes is formulated. An approximate solution algorithm based on submodular maximization is proposed and its performance is investigated. Beside WDNs, WSN applications for smart cities share a common characteristic: the area to monitor usually has a network structure. Therefore, the studies of this thesis can be potentially generalized for several IoT scenarios.

QC 20160419

Quality of municipal water is sustained by addition of disinfectant, generally chlorine, to the water distribution network. Because of health problems, chlorine concentration in the network is limited between maximum and minimum limits. Cancerogenic disinfectant by-products start to occur at high concentrations so it is desired to have minimum amount of chlorine without violating the limit. In addition to the health issues, minimum injection amount is favorable concerning cost. Hence, an optimization model is necessary which covers all of these considerations. However, there are uncertain factors as chlorine is reactive and decays both over time and space. Thus, probabilistic approach is necessary to obtain reliable and realistic results from the model. In this study, a linear programming model is developed for the chance constrained optimization of the water distribution network. The objective is to obtain minimum amount of injection mass subjected to maintaining more uniformly distributed chlorine concentrations within the limits while considering the randomness of chlorine concentration by probability distributions. Network hydraulics and chlorine concentration computations are done by the network simulation software, EPANET.

Drinking water distribution networks such as South East Water Ltd. (SEWL), Melbourne Water, Sydney Water, etc. are supposed to transport only dissolved matter rather than a few visible particles. However, it is almost impossible to make the drinking water free from suspended solid particles. The ability to determine the origins of these particles varies between different water supply systems, with possible sources being from catchment, treatment processes, biofilm growth within the water supply pipes, and corrosion products. Improvement of our understanding of the complex hydrodynamic behavior of suspended and/or deposited particles involved in these distribution pipe networks requires mathematical and physical models. Computational Fluid Dynamics (CFD) along with analytical turbulent model is one of the most popular mathematical techniques, which has the ability to predict the behavior of complex flows for such multiphase flow applications. This study has been completed mainly in two steps. A CFD investigation was carried out to predict the hydrodynamic behavior of turbid particle flowing through a horizontal pipe networks including loop consist of bends and straight pipes. Furthermore, an extended analytical model was re-developed for the liquid-solid system to look at the similar behavior of the solid particles flowing in a turbulent field. These two parallel studies will provide better understandings about the turbidity spikes movements in the distribution networks. A comprehensive CFD investigation was carried out for particle deposition in a horizontal pipe loop consisting of four 900 bends in a turbulent flow field. A satisfactory agreement was established with the experimental data as validation. This was a steady state multi-particle problem, which helped to understand the deposition characteristics for different particle sizes and densities at upstream and downstream sides of the bends as well as its circumference. Particle concentration was seen high at the bottom wall in the pipe flow before entering the bends, but for the downstream of bend the deposition was not seen high at the bottom as seen in upstream of bend rather inner side of the bend wall (600 skewed from bottom). The larger particles clearly showed deposition near the bottom of the wall except downstream. As expected, the smaller particles showed less tendency of deposition and this was more pronounced at higher velocity. Due to the high stream line curvature and associated centrifugal force acting on the fluid at different depths the particles became well mixed and resulted in homogeneous distribution near the bend regions. The hydrodynamic behavior of particles flowing in a turbulent unsteady state flowing through a horizontal pipe was also studied to compare with the drinking water distribution networks data. In this numerical simulation six different flow-profiles and particle-load profiles were used to compute particles deposition and re-entrainment into the systems and to identify the conditions of the deposition and suspension mechanisms. Results showed that after a certain length of pipe and period of time after downward velocity gradient, when the velocity was constants over time, the shear stress was sufficiently high enough to cause the particle deposition on and roll along the bottom wall of pipe wall and created a secondary group of particle peak (called kink). Finally, an extended analytical Turbulent Diffusion Model for liquid-solid phase was developed following an existing gas-liquid turbulence model. This turbulent diffusion model was then compared with the results of the CFD investigation making use of the same boundary conditions. The comparison established good agreement between these two models. The influence of velocity on the particle size distribution was dominant over the influence of the superficial liquid velocity, which was also explained by using the new parameter, velocity ratio. This velocity ratio was defined as the ratio between the free flight and gravitational velocity. Due to some inevitable assumptions used in the analytical model, the results showed typically less deposition as compared with the CFD investigation.

This work concerns the concept of on-line monitoring and control for water distribution networks. The problem is simple to state. It is to produce a robust scheme that can continuously provide reliable information about the state of a water network in real-time and over extended periods with the minimum of operator interaction. This thesis begins by proposing a relational database schema to store 'asset data' for a water distribution network and asserts that asset data should be used as a basis for network modelling. It presents a topology determination algorithm and a demand allocation algorithm so that a mathematical model can be maintained on-line, with operator intervention only necessary to record the change of state of non-telemetered plant items such as switch valves. In order to provide a reliable on-line model of a distribution system, an investigation has been carried out into the methods available for modelling water networks and in particular, the inherent assumptions in these practices. As a result, new methods have been produced for network element combination and for demand allocation. These methods both support the database approach and enhance the robustness of the system by increasing the range of conditions for which the resulting model is applicable. For operational control, a new technique for state estimation is proposed which combines the advantages of weighted least squares estimation with those of weighted least absolute values estimation. The proposed method is tolerant to transducer noise and to the presence of large measurement outliers. However, the method is not limited in its application to water networks and could be applied to a wide range of measurement processing problems. Lastiy, a new topology based method for processing suspect data is proposed which can determine the likely causes using identifying templates. Thus a new approach to water network monitoring is proposed via an overall framework into which the various tasks of on-line operational control can be integrated. The exercise has resulted in the production of a core software package which could realistically be used in a control room to facilitate reliable operational control of water distribution systems.

The eective management of water distribution systems is gaining tremendousfocus within the scientic community thanks to the Internet of Things.This report is an exploration attempt into the mathematical optimization ofa water distribution network under the inuence of leakages caused by deteriorationwithin the distribution network. The optimization process is carriedout in a two phase manner involving the optimal placement of pressure valves,followed by the optimal control of the valves in the presence of known leakswithin the network. The optimization framework addresses the minimisationof the average network pressure in an extended time setting by imposingthe hydraulic equations as nonlinear constraints. The hydraulic components,namely the pressure reduction valves, are modelled as integer variables leadingto a non-convex and non-linear optimization problem that falls under theclass of optimization problems known as mixed-integer nonlinear programming(MINLP). The exercise implements two reformulation methods that solves theMINLP problem as a sequence of regular nonlinear programs (NLPs), and alsopresents the hydraulic simulation results of the implementation. While thereis sucient research on water network optimization using various mathematicalmethods, this study endeavours to combine a leakage model within theoptimization framework and presents the ndings of the analysis. In addition,the report also includes the outcomes of the simulation on a real distributionnetwork simulated under varying demand conditions.
Effektiv förvaltning av vattendistributionssystem har skaffat stor inriktning inom det vetenskapliga samhället på grund av Sakernas internet (engelska IoT). En matematisk optimering av ett vattendistributionsnät som påverkades av läckage på grund av försämring undersöktes i den här rapporten. Optimeringsprocessen utfördes i tvåstegs med den optimala placeringen av tryckventiler i första steget följt av optimal styrning av ventilerna i närvaro av kända läckor inom nätverket. Optimeringsramen behandlar minimering av det genomsnittliga nätverkstrycket i en förlängd tidsinställning genom att införa hydrauliska ekvationer som icke-linjära begränsningar. De hydrauliska komponenterna, nämligen tryckreduceringsventilerna, modellerades som integervariabler somleder till ett icke-konvex och icke-linjärt optimeringsproblem som kallas mixedinteger icke-linjära programmering (engelska MINLP). Ö vningen genomförtvå reformuleringsmetoder som löser MINLP-problemet som regelbundna ickelinjära program (NLP) och presenterar också hydrauliska simuleringsresultatav det. Ä ven om det finns tillräcklig forskning om optimering av vattennätverkmed hjälp av olika matematiska metoder strävar det här arbetet efter att kombinera en hydraulisk läckagemodell inom nätverkoptimeringsramen och presenterar analysens resultat. Dessutom innehåller rapporten även simuleringsresultaten på ett riktigt distributionsnät simulerad under varierande vat-tenefterfrågan.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2011.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 34-35).
Human society and civilization rely on the constant availability of fresh water. In regions where a local source of potable water is not available, a transportation and distribution pipe system is employed. When these pipes feature cracks, holes, or leaks, the result is a substantial waste of energy and natural resources. As communities grow the loss due to these flaws becomes more costly, and the motivation to detect leaks increases. The purpose of this thesis project is to develop pressure difference-based sensing cells that can be used in an untethered leak-detection device. This device is to be deployed in water distribution networks to locate leaks so that water loss can be minimized. Design of these sensing cells and of the leak-detection device entails evaluating the size and shape of a leak's low pressure region. In this paper, leaks are investigated in this regard and a number of different pressure difference-sensing sensor technologies are explored and evaluated. A silicone-rubber deflecting membrane is selected for the application. The relationship between pressure-derived force acting on its surface and its maximum deflection is evaluated as a means of leak detection. Ultimately, testing reveals that these types of cells are simple and robust. While they deflect as anticipated, the formula used to predict their behavior does not fit the experimental results. It is concluded that this type of pressure difference-sensing membrane is well-suited for application within an untethered sensor, with the opportunity for deeper material selection and more accurate deflection analysis.
by Pall Magnus Kornmayer.
S.B.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Civil Engineering, 1985.
MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING.
Bibliography: leaf 95.
by Lisa A. Jeffrey.
M.S.

Il tema della sostenibilità e del conseguente utilizzo di nuove tecniche per la gestione delle risorse idriche è da sempre molto complesso. La digitalizzazione e l'impiego di nuove tecnologie in campo idrico ha ridotto sia i costi che le tempistiche di molteplici processi. Un valido esempio è costituito dalla modellazione surrogata: Il tema centrale della tesi ha come scopo la costruzione di un modello surrogato di rete di distribuzione idrica utilizzando i principi di Complex Network Theory. In particolare, per condurre questo tipo di studio è stato scelto di analizzare da un punto di vista interdisciplinare i concetti di robustezza e di resilienza ai malfunzionamenti. Le reti di distribuzione idrica possono essere facilmente assimilabili alle reti complesse presenti nella teoria dei grafi. Gli elementi strutturali delle reti idriche sono convertiti nell'equivalente grafo complesso costituito da archi e nodi. Al fine di indagare una possibile correlazione tra le reti di entrambe le discipline, sono state eseguite una serie di simulazioni idriche e valutazioni strutturali calibrando i modelli in base alle euristiche proposte in letteratura. Gli esperimenti eseguiti valutano l'effetto "small world" e altre misure per l'analisi di reti presenti nella teoria dei grafi in relazione all'intensità e alla durata dei malfunzionamenti. Dai risultati è stata riscontrata una correlazione inversa tra la proprietà di "small world" e l'indice di durata dei malfunzionamenti. In aggiunta, i risultati ottenuti evidenziano una tendenza generale alla correlazione, sia diretta che indiretta. Un ulteriore evidenza emersa dall'analisi è la doppia correlazione presente tra altezza dei nodi, durata dei malfunzionamenti e indice di centralità per vicinanza. Questo lavoro può dunque aiutare nell'individuazione di parametri più puntuali e adeguati suggerendo nuove vie e tecniche per una ancora più corretta creazione di modelli surrogati tramite tecniche di Complex Network.

Aging infrastructure and personnel, combined with stricter financial constraints has put maintenance, or more popular Asset Management, at the top of the agenda for most power utilities. At the same time the industry reports that this area is not properly supported by information systems. Today’s power utilities have very comprehensive and complex portfolios of information systems that serve many different purposes. A common problem in such heterogeneous system architectures is data management, e.g. data in the systems do not represent the true status of the equipment in the power grid or several sources of data are contradictory. The research presented in this thesis concerns how this industrial problem can be better understood and approached by novel use of the ontology standardized in the Common Information Model defined in IEC standards 61970 & 61968. The theoretical framework for the research is that of data management using ontology based frameworks. This notion is not new, but is receiving renewed attention due to emerging technologies, e.g. Service Oriented Architectures, that support implementation of such ontological frameworks. The work presented is empirical in nature and takes its origin in the ontology available in the Common Information Model. The scope of the research is the applicability of the CIM ontology, not as it was intended i.e. in systems integration, but for analysis of business processes, legacy systems and data. The work has involved significant interaction with power distribution utilities in Sweden, in order to validate the framework developed around the CIM ontology. Results from the research have been published continuously, this thesis consists of an introduction and summary and papers describing the main contribution of the work. The main contribution of the work presented in this thesis is the validation of the proposition to use the CIM ontology as a basis for analysis existing legacy systems. By using the data models defined in the standards and combining them with established modeling techniques we propose a framework for information system management. The framework is appropriate for analyzing data quality problems related to power systems maintenance at power distribution utilities. As part of validating the results, the proposed framework has been applied in a case study involving medium voltage overhead line inspection. In addition to the main contribution, a classification of the state of the practice system support for power system maintenance at utilities has been created. Second, the work includes an analysis and classification of how high performance Wide Area communication technologies can be used to improve power system maintenance including improving data quality.
QC 20100614

A completely satisfactory water distribution network should fulfill its basic requirements such as providing the expected quality and quantity of water with the desired residual pressures during its lifetime. A water distribution network should accommodate the abnormal conditions caused by failures. These types of failures can be classified into two groups
mechanical failures and hydraulic failures. Mechanical failure is caused due to malfunctioning of the network elements such as pipe breakage, power outage and pump failure. On the other hand, hydraulic failure, considers system failure due to distributed flow and pressure head which are inadequate at one or more demand points.This study deals with the calculation of the hydraulic system reliability of an existing water distribution network regarding the Modified Chandapillai model while calculating the partially satisfied nodes. A case study was carried out on a part of Ankara Water Distribution Network, N8-1. After the modeling of the network, skeletonization and determination of nodal service areas were carried out. The daily demand curves for the area were drawn using the data that were taken from SCADA of the water utility. The daily demand curves of different days were joined and one representative mean daily demand curve together with the standard deviation values was obtained. The friction coefficient values of the pipes and storage tank water elevation were taken as other uncertainty parameters for the model. Bao and Mays (1990) approach were carried together with the hydraulic network solver program prepared by Nohutcu (2002) based on Modified Chandapillai model. The sensitivity analysis for the effects of system characteristics and model assumptions were carried out to see the effects of the parameters on the calculations and to investigate the way of improving the hydraulic reliability of the network. The storage tank should be located at a higher level for improving the reliability of the network. Also having the storage tank water level nearly full level helps in improving the reliability in daily management. Moreover, the hydraulic system reliability is highly dependent on the pumps as the lowest reliability factors were the ones with the no pump scenarios. Determining the required pressures for nodes are very important since they are the dominant factors that effects the reliability calculations. On the other hand, friction coefficient parameters and type of probability distribution function do not have dominant effect on the results. Results of this study were helpful to see the effects of different parameters on the hydraulic reliability calculations and for assessment of the methods for improving the reliability for the network.

The detection of bacteria in the water is a slow process that requires the use of expensive equipment and qualied personnel. However, real time fast detection is essential in water distribution networks. In this thesis we study the deployment of a wireless network of biosensors in a water distribution system, in order to detect contamination of a particular kind of harmful bacteria, the E.coli. This network will e-ciently utilize the interconnected biosensors and achieve real time and in-eld detection of the bacteria. Because of the non existence of biosensors hardware equipped with radio receivers and transmitters, we study theoretically the modeling of such a system and its potential application in real water distribution networks. The main goal of our study is to nd an optimal sensor placement strategy to maximize the probability of detection, having a xed number of sensors that must be placed in a connected topology. We propose a simple algorithm that solves the optimal sensor placement problem. The performance of the proposed approach have been evaluated by considering three dierent topologies simulated by the system simulator EPANET. The simulation results show that the proposed algorithm provides the higher detection probability in the network compared to other solutions, such as random sensor placement.

This PhD thesis presents a methodology to detect, estimate and localize water leaks (with the main focus in the localization problem) in water distribution networks using hydraulic models and machine learning techniques. The actual state of the art is introduced, the theoretical basis of the machine learning techniques applied are explained and the hydraulic model is also detailed. The whole methodology is presented and tested into different water distribution networks and district metered areas based on simulated and real case studies and compared with published methods. The focus of the contributions is to bring more robust methods against the uncertainties that effects the problem of leak detection, by dealing with them using the self-similarity to create features monitored by the change detection technique intersection-of-confidence-interval, and the leak localization where the problem is tackled using machine learning techniques. By using those techniques, it is expected to learn the leak behavior considering their uncertainty to be used in the diagnosis stage after the training phase. One method for the leak detection problem is presented that is able to estimate the leak size and the time that the leak has been produced. This method captures the normal, leak-free, behavior and contrast it with the new measurements in order to evaluate the state of the network. If the behavior is not normal check if it is due to a leak. To have a more robust leak detection method, a specific validation is designed to operate specifically with leaks and in the temporal region where the leak is most apparent. A methodology to extent the current model-based approach to localize water leaks by means of classifiers is proposed where the non-parametric k-nearest neighbors classifier and the parametric multi-class Bayesian classifier are proposed. A new data-driven approach to localize leaks using a multivariate regression technique without the use of hydraulic models is also introduced. This method presents a clear benefit over the model-based technique by removing the need of the hydraulic model despite of the topological information is still required. Also, the information of the expected leaks is not required since information of the expected hydraulic behavior with leak is exploited to find the place where the leak is more suitable. This method has a good performance in practice, but is very sensitive to the number of sensor in the network and their sensor placement. The proposed sensor placement techniques reduce the computational load required to take into account the amount of data needed to model the uncertainty compared with other optimization approaches while are designed to work with the leak localization problem. More precisely, the proposed hybrid feature selection technique for sensor placement is able to work with any method that can be evaluated with confusion matrix and still being specialized for the leak localization task. This last method is good for a few sensors, but lacks of precision when the number of sensors to place is large. To overcome this problem an incremental sensor placement is proposed which is better for a larger number of sensors to place but worse when the number is small.
Aquesta tesi presenta una nova metodologia per a localització de fuites en xarxes de distribució d'aigua potable. Primer s'ha revisat l'estat del art actual i les bases teòriques tant de les tècniques de machine learning utilitzades al llarg de la tesi com els mètodes existents de localització de fuites. La metodologia presentada s'ha provat en diferents xarxes d'aigua simulades i reals, comparant el resultats amb altres mètodes publicats. L'objectiu principal de la contribució aportada és el de desenvolupar mètodes més robustos enfront les incerteses que afecten a la localització de fuites. En el cas de la detecció i estimació de la magnitud de la fuita, s'utilitza la tècnica self-similarity per crear els indicadors es monitoritzen amb la tècnica de detecció de canvis ("intersection-of-confidence-intervals"). En el cas de la localització de les fuites, s'han fet servir les tècniques de classificadors i interpoladors provinents del machine learning. A l'utilitzar aquestes tècniques s'espera captar el comportament de la fuita i de la incertesa per aprendre i tenir-ho en compte en la fase de la localització de la fuita. El mètode de la detecció de fallades proposat és capaç d'estimar la magnitud de la fuita i l'instant en que s'ha produït. Aquest mètode captura el comportament normal, sense fuita, i el contrasta amb les noves mesures per avaluar l'estat de la xarxa. En el cas que el comportament no sigui el normal, es procedeix a comprovar si això és degut a una fuita. Per tenir una mètode de detecció més robust, es fa servir una capa de validació especialment dissenyada per treballar específicament amb fuites i en la regió temporal en que la fuita és més evident. Per tal de millorar l'actual metodologia de localització de fuites mitjançant models hidràulics s'ha proposat l'ús de classificadors. Per una banda es proposa el classificador no paramètric k-nearest neighbors i per l'altre banda el classificador Bayesià paramètric per múltiples classes. Finalment, s'ha desenvolupat un nou mètode de localització de fuites basat en models de dades mitjançant la regressió de múltiples paràmetres sense l'ús del model hidràulic de la xarxa. Finalment, s'ha tractat el problema de la col·locació de sensors. El rendiment de la localització de fuites està relacionada amb la col·locació de sensors i és particular per a cada mètode de localització. Amb l'objectiu de maximitzar el rendiment dels mètodes de localització de fuites presentats anteriorment, es presenten i avaluen tècniques de col·locació de sensors específicament dissenyats ja que el problema de combinatòria no es pot manejar intentant cada possible combinació de sensors a part de les xarxes més petites amb pocs sensors per instal·lar. Aquestes tècniques de col·locació de sensors exploten el potencial de les tècniques de selecció de variables per tal de realitzar la tasca desitjada.
Esta tesis doctoral presenta una nueva metodología para detectar, estimar el tamaño y localizar fugas de agua (donde el foco principal está puesto en el problema de la localización de fugas) en redes de distribución de agua potable. La tesis presenta una revisión del estado actual y las bases de las técnicas de machine learning que se aplican, así como una explicación del modelo hidráulico de las redes de agua. El conjunto de la metodología se presenta y prueba en diferentes redes de distribución de agua y sectores de consumo con casos de estudio simulados y reales, y se compara con otros métodos ya publicados. La contribución principal es la de desarrollar métodos más robustos frente a la incertidumbre de los datos. En la detección de fugas, la incertidumbre se trata con la técnica del self-similarity para la generación de indicadores que luego son monitoreados per la técnica de detección de cambios conocida como intersection-of-confidece-interval. En la localización de fugas el problema de la incertidumbre se trata con técnicas de machine learning. Al utilizar estas técnicas se espera aprender el comportamiento de la fuga y su incertidumbre asociada para tenerlo en cuenta en la fase de diagnóstico. El método presentado para la detección de fugas tiene la habilidad de estimar la magnitud y el instante en que la fuga se ha producido. Este método captura el comportamiento normal, sin fugas, del sistema y lo contrasta con las nuevas medidas para evaluar el estado actual de la red. En el caso de que el comportamiento no sea el normal, se comprueba si es debido a la presencia de una fuga en el sistema. Para obtener un método de detección más robusto, se considera una capa de validación especialmente diseñada para trabajar específicamente con fugas y durante el periodo temporal donde la fuga es más evidente. Esta técnica se compara con otras ya publicadas proporcionando una detección más fiable, especialmente en el caso de fugas pequeñas, al mismo tiempo que proporciona más información que puede ser usada en la fase de la localización de la fuga permitiendo mejorarla. El principal problema es que el método es más lento que los otros métodos analizados. Con el fin de mejorar la actual metodología de localización de fugas mediante modelos hidráulicos, se propone la utilización de clasificadores. Concretamente, se propone el clasificador no paramétrico k-nearest neighbors y el clasificador Bayesiano paramétrico para múltiples clases. La propuesta de localización de fugas mediante modelos hidráulicos y clasificadores permite gestionar la incertidumbre de los datos mejor para obtener un diagnóstico de la localización de la fuga más preciso. El principal inconveniente recae en el coste computacional, aunque no se realiza en tiempo real, de los datos necesarios por el clasificador para aprender correctamente la dispersión de los datos. Además, el método es muy dependiente de la calidad del modelo hidráulico de la red. En el campo de la localización de fugas, se a propuesto un nuevo método de localización de fugas basado en modelos de datos mediante la regresión de múltiples parámetros sin el uso de modelo hidráulico. Este método presenta un claro beneficio respecto a las técnicas basadas en modelos hidráulicos ya que prescinde de su uso, aunque la información topológica de la red es aún necesaria. Además, la información del comportamiento de la red para cada fuga no es necesario, ya que el conocimiento del efecto hidráulico de una fuga en un determinado punto de la red es utilizado para la localización. Este método ha dado muy buenos resultados en la práctica, aunque es muy sensible al número de sensores y a su colocación en la red. Finalmente, se trata el problema de la colocación de sensores. El desempeño de la localización de fugas está ligado a la colocación de los sensores y es particular para cada método. Con el objetivo de maximizar el desempeño de los métodos de localización de fugas presentados, técnicas de colocación de sensores específicamente diseñados para ellos se han presentado y evaluado. Dado que el problema de combinatoria que presenta no puede ser tratado analizando todas las posibles combinaciones de sensores excepto en las redes más pequeñas con unos pocos sensores para instalar. Estas técnicas de colocación de sensores explotan el potencial de las técnicas de selección de variables para realizar la tarea deseada. Las técnicas de colocación de sensores propuestas reducen la carga computacional, requerida para tener en cuenta todos los datos necesarios para modelar bien la incertidumbre, comparado con otras propuestas de optimización al mismo tiempo que están diseñadas para trabajar en la tarea de la localización de fugas. Más concretamente, la propuesta basada en la técnica híbrida de selección de variables para la colocación de sensores es capaz de trabajar con cualquier técnica de localización de fugas que se pueda evaluar con la matriz de confusión y ser a la vez óptimo. Este método es muy bueno para la colocación de sensores, pero el rendimiento disminuye a medida que el número de sensores a colocar crece. Para evitar este problema, se propone método de colocación de sensores de forma incremental que presenta un mejor rendimiento para un número alto de sensores a colocar, aunque no es tan eficaz con pocos sensores a colocar.

A water distribution network analysis method known as the gradient method, due to Todini (1979), has been generalised and subjected to an extensive program of testing and evaluation. The method has been extended to include pumps and some pressure regulating valves, and an original physically-based method has been proposed for modelling the latter devices. Also, a generalised version of the algorithm which considers the nodal demands as a linear function of the pressures has been introduced. The gradient method has been tested with numerous examples, showing remarkable robustness and convergence speed when compared with the most efficient traditional methods. The gradient algorithm does not break down with disconnected networks. The performance of the gradient algorithm when using seven different linear solvers, including direct and iterative methods, has been investigated. A multifrontal linear solver has been identified as the most efficient method when enough computer memory is available (routine MA27 of the Harwell Library); if storage is limited, a preconditioned (modified) conjugate gradient method is the recommended linear solver. A good compromise between memory and speed is represented by the one-way dissection method of George and Liu (1981). An automatic calibration algorithm has been proposed which estimates the true pipe resistance parameters, based on the estimation of the unmeasured piezometric heads and unmeasured flows. For the piezometric head estimation, three different methods have been proposed and compared: one based on Kriging, another based on bi-cubic splines and a third based on an original deterministic one-dimensional interpolation procedure. The latter producing the closest estimates with respect to the true values. For the estimation of the unmeasured flows, the raw (un-calibrated) network model itself is used, based on initial estimates of the pipe roughnesses, leading to an iterative procedure. The results of using the proposed calibration algorithm with a set of test examples show that the unmeasured flow estimation needs further work and an alternative approach has been suggested, which, hopefully, would lead to improvements both in the flow estimation and in the estimation of the true roughnesses.

Les réseaux de distribution d’eau consistent en de grandes infrastructures réparties dans l’espace qui assurent la distribution d’eau potable en quantité et en qualité suffisantes. Les modèles mathématiques de ces systèmes sont caractérisés par un grand nombre de variables d’état et de paramètres dont la plupart sont incertains. Les temps de calcul peuvent s’avérer conséquents pour les réseaux de taille importante et la propagation d’incertitude par des méthodes de Monte Carlo. Par conséquent, les deux principaux objectifs de cette thèse sont l’étude des techniques de modélisation à ordre réduit par projection ainsi que la propagation spectrale des incertitudes des paramètres. La thèse donne tout d’abord un aperçu des méthodes mathématiques utilisées. Ensuite, les équations permanentes des réseaux hydrauliques sont présentées et une nouvelle méthode de calcul des sensibilités est dérivée sur la base de la méthode adjointe. Les objectifs spécifiques du développement de modèles d’ordre réduit sont l’application de méthodes basées sur la projection, le développement de stratégies d’échantillonnage adaptatives plus efficaces et l’utilisation de méthodes d’hyper-réduction pour l’évaluation rapide des termes résiduels non linéaires. Pour la propagation des incertitudes, des méthodes spectrales sont introduites dans le modèle hydraulique et un modèle hydraulique intrusif est formulé. Dans le but d’une analyse plus efficace des incertitudes des paramètres, la propagation spectrale est ensuite évaluée sur la base du modèle réduit. Les résultats montrent que les modèles d’ordre réduit basés sur des projections offrent un avantage considérable par rapport à l’effort de calcul. Bien que l’utilisation de l’échantillonnage adaptatif permette une utilisation plus efficace des états système pré-calculés, l’utilisation de méthodes d’hyper-réduction n’a pas permis d’améliorer la charge de calcul. La propagation des incertitudes des paramètres sur la base des méthodes spectrales est comparable aux simulations de Monte Carlo en termes de précision, tout en réduisant considérablement l’effort de calcul
Water distribution systems are large, spatially distributed infrastructures that ensure the distribution of potable water of sufficient quantity and quality. Mathematical models of these systems are characterized by a large number of state variables and parameter. Two major challenges are given by the time constraints for the solution and the uncertain character of the model parameters. The main objectives of this thesis are thus the investigation of projection based reduced order modelling techniques for the time efficient solution of the hydraulic system as well as the spectral propagation of parameter uncertainties for the improved quantification of uncertainties. The thesis gives an overview of the mathematical methods that are being used. This is followed by the definition and discussion of the hydraulic network model, for which a new method for the derivation of the sensitivities is presented based on the adjoint method. The specific objectives for the development of reduced order models are the application of projection based methods, the development of more efficient adaptive sampling strategies and the use of hyper-reduction methods for the fast evaluation of non-linear residual terms. For the propagation of uncertainties spectral methods are introduced to the hydraulic model and an intrusive hydraulic model is formulated. With the objective of a more efficient analysis of the parameter uncertainties, the spectral propagation is then evaluated on the basis of the reduced model. The results show that projection based reduced order models give a considerable benefit with respect to the computational effort. While the use of adaptive sampling resulted in a more efficient use of pre-calculated system states, the use of hyper-reduction methods could not improve the computational burden and has to be explored further. The propagation of the parameter uncertainties on the basis of the spectral methods is shown to be comparable to Monte Carlo simulations in accuracy, while significantly reducing the computational effort

Leaks in pipe networks cause significant problems for utilities and water users in terms of lost revenue and interrupted service. In many cities the leakage is as high as forty percent. A water audit is carried out to assess system-wide leakage. However, to detect leakage at the level of a pipeline, a physical measurement technique is generally employed. For large cities the distribution piping length amounts to a few thousand miles. Therefore, the physical measurements can become tedious and expensive. In this thesis it is assumed that a spatial distribution of leakage can be estimated at nodes based on a water audit bookkeeping scheme. A mathematical formulation consisting of continuity, energy (headloss), pressure-dependent demands and/or leakage, and flow direction preservation equations are utilized to distribute demand flows and leakage among pipes. The leakage is attributed to the formation of corrosion holes. Based upon the extent of corrosion, the leakage flow arriving at a particular node is apportioned among all pipes that are converging at that node. Therefore, the formulation presented in this thesis captures the two essential elements behind leakage, namely, pressure driven flow distribution and the vulnerability of pipes to corrosion. The proposed formulation allows utilities to be more proactive in identifying leakage prone districts within the water distribution system. An understanding of the pressure-dependent leakage in the system is helpful when performing a water audit and in developing strategies for leak repair programs. Restoring the full capacity of the water distribution system will greatly increase the reliability of the system, thereby benefiting local utilities and water users.
Master of Science

Nitrification in water distribution networks has become a growing concern for water supplies in the United States. The use of chloramines as a disinfectant in distribution pipe networks has become increasingly popular to reduce the disinfectant byproducts that are formed with free chlorine. In chloraminated systems there is potential for nitrification to occur because it reduces chloramine residuals. As chloramines decompose in the network, ammonia is released. Nitrifiers oxidize ammonia into nitrites, which react with chloramines resulting in its further reduction. As this cycle continues, chloramines will be consumed faster in the network, causing regrowth of heterotrophic bacteria. A study was conducted to compare the Fargo and Moorhead distribution networks for the occurrence of nitrification and their potential to deteriorate water quality. Each distribution network was analyzed independently for variations in operational conditions and water quality parameters that can serve as indications of nitrification in a distribution network.
Moorhead Water Treatment Plant

Much of the world's rapidly growing urban population relies upon water distribution systems to provide treated water through networks of pipes. Rather than continuously supplying water to users, many of these distribution systems operate intermittently, with parts of the network frequently losing pressure or emptying altogether. Such intermittent water supply deleteriously impacts water availability, infrastructure, and water quality for hundreds of millions of people around the world. In this work I introduce the problem of intermittent water supply through the lens of applied mathematics. I first introduce a simple descriptive mathematical model that captures some hydraulic features of intermittency not accounted for by existing water distribution system software packages. I then consider the potential uses of such a model in a variety of optimization examples motivated by real-world applications. In simple test networks, I show how to reduce pressure gradients while the network fills by changing either the inflow patterns or the elevation profile. I also show test examples of using measured data to potentially recover unknown information such as initial conditions or boundary outflows. I then use sensitivity analysis to better understand how various parameters control model output, with an eye to figuring out which parameters are worth measuring most carefully in field applications, and also which parameters may be useful in an optimization setting. I lastly demonstrate some progress in descriptively modeling a real network, both through the introduced mathematical model and through a fluid-mechanics-based method for identifying data where the model is most useful.