This thesis is concerned with the developments of analysis, modelling and optimization techniques and computer program algorithms, with the ultimate aim of control of water supply and distribution systems to lead to overall optimal operation. Typical system features and operational conditions are analyzed, and the requirements for the overall objective are examined, to determine an overall control strategy which is subsequently developed and tested on real systems throughout this thesis. As a prerequisite, short-term water demand forecasting is extensively studied by employing time series analysis. Special consideration is given to improving the forecasting accuracy of the method and its on-line implementation. In order to speed up the solution time of optimal system operation, simplified system models -- namely, piecewise macroscopic model and equivalent network model -- are developed respectively. Then by employing the piecewise macroscopic model, a nonlinear programming method is developed to cater for the optimal operation of a class of multi-source systems without significant storage. The optimal operation policy obtained by this method is realized at two levels: the first level calculates the optimized apportioning of water to be delivered by different sources; the second level decides the least cost pump schedules to supply the optimized apportioning of water. Based on the equivalent network model, a linear programming method is developed for optimization of a class of multi-source, multi-reservoir systems with a mixture of fixed speed pumps and variable speed and/or variable throttle pumps. This method yields directly optimized pump schedules and reservoir trajectories in terms of least cost system operation. The integration of the developments results in a scheme which can be applied to give overall dynamic control of a wide range of water supply and distribution systems. The application results presented in this thesis justify the theoretical developments and show that benefits can be obtained from these developments.

Thesis (MLHLAL)--University of Michigan, 1978.

Thesis (MLHLAL)--University of Michigan, 1978.

A comprehensive study on the application of flow cytometry (FCM) for the analysis of biofilms has been undertaken and the results presented in this thesis have shown that flow cytometry can been successfully used to enumerate, sort and image the bacteria and amoebae in biofilms and water distribution systems as a rapid and sensitive semiautomated technique compared with conventional microbiology. It has been shown that the results of flow cytometric analysis of total Legionella pneumophila cells have a strong statistical correlation with the numbers of Legionella cfu by BCYE plate counting (BCYE PC) methods for biofilms and planktonic phases. There are also strong statistical correlations between flow cytometric analysis and epifluorescent microscopic (EFM) analysis (direct counting) for determination of bacteria, including Legionella, Escherichia coli, Salmonella, Pseudomonas and amoebae, and total and viable cells in pure cultures, water distribution systems and biofilms. The flow cytometric protocols have been set up and optimised for the analysis of environmental microorganisms. The novel fluorescent dyes and immunofluorescence antibodies from the most current commercial dyes also have been screened and the staining protocols have been optimised and adopted for flow cytometric analysis and direct counting by epifluorescent microscopy. The tap water biofilms and river water biofilms were analysed by the flow cytometer and direct counting methods as well as by conventional microbiological methods (plate counting). The bacterial populations in real water distribution systems have been fully investigated and the total, viable bacteria were determined by the above methods. The findings of this work have practical implications with respect to the rapid and automatic detection and predictions of Legionella spp. and the risk assessment from biofilms and water environments.

Multi-objective design or extended design of Water Distribution Systems (WDSs) has received more attention in recent years. It is of particular interest for obtaining the trade-offs between cost and hydraulic benefit to support the decision-making process. The design problem is usually formulated as a multi-objective optimisation problem, featuring a huge search space associated with a great number of constraints. Multi-objective evolutionary algorithms (MOEAs) are popular tools for addressing this kind of problem because they are capable of approximating the Pareto-optimal front effectively in a single run. However, these methods are often held by the “No Free Lunch” theorem (Wolpert and Macready 1997) that there is no guarantee that they can perform well on a wide range of cases. To overcome this drawback, many hybrid optimisation methods have been proposed to take advantage of multiple search mechanisms which can synergistically facilitate optimisation. In this thesis, a novel hybrid algorithm, called Genetically Adaptive Leaping Algorithm for approXimation and diversitY (GALAXY), is proposed. It is a dedicated optimiser for solving the discrete two-objective design or extended design of WDSs, minimising the total cost and maximising the network resilience, which is a surrogate indicator of hydraulic benefit. GALAXY is developed using the general framework of MOEAs with substantial improvements and modifications tailored for WDS design. It features a generational framework, a hybrid use of the traditional Pareto-dominance and the epsilon-dominance concepts, an integer coding scheme, and six search operators organised in a high-level teamwork hybrid paradigm. In addition, several important strategies are implemented within GALAXY, including the genetically adaptive strategy, the global information sharing strategy, the duplicates handling strategy and the hybrid replacement strategy. One great advantage of GALAXY over other state-of-the-art MOEAs lies in the fact that it eliminates all the individual parameters of search operators, thus providing an effective and efficient tool to researchers and practitioners alike for dealing with real-world cases. To verify the capability of GALAXY, an archive of benchmark problems of WDS design collected from the literature is first established, ranging from small to large cases. GALAXY has been applied to solve these benchmark design problems and its achievements in terms of both ultimate and dynamic performances are compared with those obtained by two state-of-the-art hybrid algorithms and two baseline MOEAs. GALAXY generally outperforms these MOEAs according to various numerical indicators and a graphical comparison tool. For the largest problem considered in this thesis, GALAXY does not perform as well as its competitors due to the limited computational budget in terms of number of function evaluations. All the algorithms have also been applied to solve the challenging Anytown rehabilitation problem, which considers both the design and operation of a system from the extended period simulation perspective. The performance of each algorithm is sensitive to the quality of the initial population and the random seed used. GALAXY and the Pareto-dominance based MOEAs are superior to the epsilon-dominance based methods; however, there is a tie between GALAXY and the Pareto-dominance based approaches. At the end, a summary of this thesis is provided and relevant conclusions are drawn. Recommendations for future research work are also made.

This thesis presents the findings of some research carried out pertaining to economic operation and planning distribution systems. An optimal capacitor switching algorithm is developed for distribution system based on backward-forward sweep algorithm, which can assist in real-time applications. Thereafter, an optimal reconfiguration algorithm is proposed for distribution networks that seek to minimize losses by reducing the number of spanning trees in the network. The proposed algorithm provides a faster solution method and is useful for practical applications. Finally, the issues of short-term operating and long term planning of distribution networks in the presence of distributed generators is examined. An optimization framework is developed to determine the optimal locations of these distributed generator units.

This thesis presents the findings of some research carried out pertaining to economic operation and planning distribution systems. An optimal capacitor switching algorithm is developed for distribution system based on backward-forward sweep algorithm, which can assist in real-time applications. Thereafter, an optimal reconfiguration algorithm is proposed for distribution networks that seek to minimize losses by reducing the number of spanning trees in the network. The proposed algorithm provides a faster solution method and is useful for practical applications. Finally, the issues of short-term operating and long term planning of distribution networks in the presence of distributed generators is examined. An optimization framework is developed to determine the optimal locations of these distributed generator units.

The future of electric power distribution systems (DSs) is one that incorporates extensive amounts of advanced metering, distribution automation, and distributed generation technologies. Most DSs were designed to be radial systems and the major philosophies of their protection, namely, selectivity and sensitivity, were easily achieved. Settings for overcurrent protective devices (OCPDs) were static and based on the maximum load downstream of its location, with little concern of major configuration changes. However, the integration of distribution generators (DGs) in radial distributions systems (RDSs) causes bidirectional power flows and varying short circuit currents to be sensed by protective devices, thereby affecting these established protection principles. Several researchers have investigated methods to preserve the selectivity of overcurrent protection coordination in RDSs with DGs, but at the expense of protective device sensitivity due to an inherent change in system configuration. This thesis presents an investigation to adapt the pickup settings of the substation relay, based on configuration changes in a DS with DGs, using smart meter data from the prior year. An existing protection scheme causes the faulted areas of DSs with DGs to revert to a radial configuration, thereby allowing conventional OCPDs to isolate faults. Based on the location of the fault, the created radial segments are known and vary in length. The proposed methodology involves using demand information available via smart metering, to determine the seasonal maximum diversified demands in each of the radial segments that are formed. These seasonal maximum diversified demands are used to yield several pickup settings for the substation overcurrent relay of the DS. The existing protection approach enables the selectivity of radial overcurrent protection coordination to be maintained; the sensitivity of the substation relay is improved by adapting its pickup settings based on seasonal demand and system configuration changes. The results of the studies are reported through simulation in EMTP™ /PSCAD® using a multi-feeder test system that includes DGs and smart meters located at the secondary distribution load level. The results show that using seasonal settings for the substation relay based on configuration changes in a DS with DGs can improve the sensitivity of the substation relay.

Impedance-based methods are used to locate faults on distribution systems because of their simplicity and ease of implementation. These methods require fault voltage and current data along with the positive- and zero-sequence line impedance values (in ohm per unit length) to estimate the reactance or distance to fault location. Inaccuracies in line impedance values, which arise from circuit model errors, have an adverse impact on fault location estimates of the impedance-based methods. Measurement errors in current and voltage transformers can also lead to inaccuracy in estimation. Further, all methods use simplistic models to represent the system load. The load in a practical distribution system does not conform to the oversimplified models leading to errors in estimation of fault location. This thesis presents sensitivity analysis of four impedance-based methods. It focuses on the Takagi, positive-sequence reactance, loop reactance and balanced-load methods. Amongst these four methods, the first three are commonly used for fault location. The fourth method was developed as a part of this work. The objective of sensitivity analysis is to study and quantify the effect of circuit model, measurement and load model errors, on the fault location estimates of the four methods. The results of this analysis are used to establish upper and lower bounds on the estimation errors for each method. The analysis begins with creation of a baseline case using a modified version of the IEEE 34 Node Test Feeder. All the methods estimate the reactance to fault location as a part of this analysis. The baseline case uses accurate line impedances and measurement values in the four methods. The fault location estimates for this case serve as a means of comparison for all subsequent analyzes. Secondly, various circuit model errors are introduced while computing the line impedance values. These errors include inaccurate modeling of four parameters viz. phase conductor distances, conductor sizes, phase to neutral conductor distances and earth resistivity. The erroneous line impedance values, which arise from these circuit model errors, are used in the four methods. The resultant location estimates are compared with those for the baseline case. It is observed that modeling errors in earth resistivity can cause estimation errors of 2% to 5% in the Takagi and positive-sequence reactance methods. These errors can be positive or negative depending upon whether the modeled earth resistivity value is more than or less than the accurate value. The effect of inaccurate modeling of the other three parameters is marginal. Additionally, the Takagi and positive-sequence reactance methods assume line impedances to be uniform while estimating fault location. Although this assumption is a type of circuit model error, it does not lead to significant errors in estimation. The loop reactance and balanced-load methods are insensitive to circuit model errors as they do not use line impedance values while estimating reactance to fault location. The next part is analysis of effect of measurement errors on fault location estimates. Ratio and phase angle errors are deliberately introduced in the current and voltage transformers and the erroneous measurements are used to conduct fault location. This causes 5% to 6% errors in estimation for the Takagi and positive-sequence reactance methods. These estimation errors can be positive or negative depending upon the magnitude of the CT and VT ratio errors and the sign of the phase angle errors. For the loop reactance method, erroneous measurements introduce 8% to 30% errors in fault location. This indicates that the loop reactance method is highly sensitive to measurement errors. The balanced-load method is moderately sensitive and experiences 6% to 7% errors in fault location estimates. Lastly, the effect of load current on fault location estimates is analyzed. When the Takagi and positive-sequence reactance methods are used on a heavily loaded system, they estimate fault location with an error of 5% to 8%. The loop reactance method is severely affected by the level of load current in the system. This method can estimate fault location with nearly 100% accuracy, on a lightly loaded system. However, the estimation errors for this method increase significantly and are in the range of 15% to 30%, as load current in the system increases. In case of the balanced-load method, unbalanced, heavy loads can cause estimation errors of 7% to 25%. The combined effect of all the error sources is taken into account by creating a confidence interval for each method. For the Takagi and positive-sequence reactance methods, the actual fault location can be expected to lie within ±10% of the estimated value. The fault location estimation error for the loop reactance and balanced-load methods is always positive. The actual reactance-to-fault is within -30% of the value estimated by these methods. / text

Silver was evaluated as disinfectant to maintain water quality in water distribution system. It was used to inhibit growth of two opportunistic bacteria in planktonik form and in biofilm formation in Robbins devices with stainless steel and PVC surfaces. The results of this work show that silver is a potential secondary disinfectant to be used in water distribution systems