Journal of Water Resources Planning and Management

An economic-engineering optimization model of California’s major water supply system is presented. The model’s development, calibration, limitations, and results are reviewed. The major methodological conclusions are that large-scale water resources optimization models driven by economic objective functions are both possible and practical; deterministic models are useful despite their limitations; and data management, reconciliation, and documentation are important benefits of large-scale system modeling. Specific results for California indicate a great potential for water markets and conjunctive use to improve economic performance and significant economic value for expanding some conveyance facilities. Overall, economic-engineering optimization (even if deterministic) can suggest a variety of promising approaches for managing large systems. These approaches can then be refined and tested using more detailed simulation models. The process of developing large-scale models also motivates the systematic and integrated treatment of surface water, groundwater, facility, and water demand data, and identification of particularly important data problems, something of long-term value for all types of water resources analysis.

Recent advances in computer technology and water resource modeling, availability of real-time hydroclimatic data, and improvements in our ability to develop user-friendly graphical model interfaces have led to significant growth in the development and application of decision support systems (DSSs) for water resource systems. This study provides an example of the development of a real-time DSS for adaptive management of the reservoir system that provides drinking water to the Boston metropolitan region. The DSS uses a systems framework to link watershed models, reservoir hydraulic models, and a reservoir water quality model with linear and nonlinear optimization algorithms. The DSS offers the ability to optimize daily and weekly reservoir operations toward four objectives based on short-term climate forecasts: (1) maximum water quality, (2) ideal flood control levels, (3) optimum reservoir balancing, and (4) maximum hydropower revenues. Case studies document the value of the DSS as an enhancement of current rule curve operations. The study shows that simple tools, in this case, familiar spreadsheet software, can be used to improve system efficiencies.

A practical monthly optimization model, called SISOPT, is developed for the management and operations of the Brazilian hydropower system. The system, one of the largest in the world, consists of 75 hydropower plants with an installed capacity of 69,375 MW, producing 92% of the nation’s electrical power. The system size and nonlinearity pose a real challenge to the modelers. The basic model is formulated in nonlinear programming (NLP). The NLP model is the most general formulation and provides a foundation for analysis by other methods. The formulated NLP model was first linearized by two different linearization techniques and solved by linear programming (LP). A comparative analysis was made of the results obtained from the linearized and the NLP models. The results show that the simplest linearized model (referred to as the LP model) without iteration is suitable for planning purposes. For example, the LP model could be used in system capacity expansion studies or to explore various design parameters in connection with feasibility studies, where details in storage variation are not as important as the power production. With a good initial policy provided by the LP model, the successive linear programming (SLP) model produced excellent results with fast convergence. The NLP model, the most complex and accurate model in the suite, is particularly suited for setting up guidelines for real-time operations using inflow forecast with frequent updating. The performance of the NLP model was checked against the historical operational records, and the comparison yields indications of superior performance.

In this paper the concept of robustness is introduced and applied to a model for the analysis of the impacts of spatially distributed policy measures on the surface water quality on a river basin scale. In this model the influence of precipitation on emissions and resuspension of pollutants in the river sediment are taken into account. The major sources of uncertainty can be found in the transport model and the emission data. The uncertainty in the prediction of future pollutant loads is dominated by the stochastic character of the precipitation. The uncertainty in the spatial representation of the emission sources seems to be of minor importance relative to the other sources of uncertainty. Evaluation of the robustness of the model conclusions on the basis of these results led to useful guidelines about how to improve the model effectively.

During the last decade, evolutionary methods such as genetic algorithms have been used extensively for the optimal design and operation of water distribution systems. More recently, ant colony optimization algorithms (ACOAs), which are evolutionary methods based on the foraging behavior of ants, have been successfully applied to a number of benchmark combinatorial optimization problems. In this paper, a formulation is developed which enables ACOAs to be used for the optimal design of water distribution systems. This formulation is applied to two benchmark water distribution system optimization problems and the results are compared with those obtained using genetic algorithms (GAs). The findings of this study indicate that ACOAs are an attractive alternative to GAs for the optimal design of water distribution systems, as they outperformed GAs for the two case studies considered both in terms of computational efficiency and their ability to find near global optimal solutions.

Shuffled Frog Leaping Algorithm (SFLA) is a meta-heuristic for solving discrete optimization problems. Here it is applied to determine optimal discrete pipe sizes for new pipe networks and for network expansions. SFLA is a population based, cooperative search metaphor inspired by natural memetics. The algorithm uses memetic evolution in the form of infection of ideas from one individual to another in a local search. The local search is similar in concept to particle swarm optimization. A shuffling strategy allows for the exchange of information between local searches to move toward a global optimum. This paper summarizes the development of SFLANET, a computer model that links SFLA and the hydraulic simulation software EPANET and its library functions. Application of SFLANET to literature network design problems is then described. Although the algorithm is in its initial stages of development, promising results were obtained.

Presented herein is an optimal design method for a multidiameter, multioutlet pipeline employing a dynamic programming approach. Optimal lengths of varying pipe sizes for a multioutlet pipeline are obtained by minimizing the annualized cost of the pipeline, which can include either fixed costs, energy costs or both. The Darcy-Weisbach formula is used to calculate friction head loss in various sections of the multioutlet pipeline. The primary input to the optimization model is the annual cost of all the available alternatives of single pipe sizes, and the output comprises the lengths of the selected diameters to form the total length of the multioutlet pipeline satisfying allowable limits on pressure variation. Applicability of the method is demonstrated through optimal design of a pipeline comprised of more than one diameter and serving as a sprinkler lateral for varying upward and downward slope conditions.

Five Austin-style sand filters were constructed by the California Department of Transportation (CALTRANS) as retrofit projects for maintenance yards and park-and-ride facilities in the Los Angeles and San Diego metropolitan areas. Each of these filter systems included storm-water monitoring equipment for collection of flow weighted composite samples. In addition, detailed records were compiled of the design elements, construction costs, and type and amount of maintenance required at each of the sites. The construction costs were relatively high because of the retrofit nature of the project and the integration of the CALTRANS storm-drain system with those of the adjoining metropolitan areas, which eliminated any opportunities for economies of scale. An analysis of performance using linear-regression techniques indicated that for sediment and almost all particle associated constituents, effluent concentration was independent of influent concentration. For instance, the average suspended solids concentration in treated runoff was 7.8±1.2 mg/L (at the 90% confidence level) regardless of observed influent concentration. The constant effluent quality produced for the particulate constituents indicates that the calculation of a percent reduction is more indicative of the influent concentration rather than the performance of the filter itself. Rejuvenation of the filter bed was required at three sites after 3 years of operation when the solids loading to the system was between 5 and 7.5 kg/m² of filter area. However, the clogging may have been accelerated by problems with the pumps that resulted in standing water on the filter for extended periods. Other maintenance activities consisted mainly of inspections, pump repair, and activities to reduce mosquito breeding. The main impediment to widespread implementation is the initial construction cost; however, modifications of the filter configuration and media may reduce these costs and increase effectiveness, thereby making the technology more attractive.