APPLICATION OF THE GENETIC ALGORITHM FOR OPTIMIZING OPERATION RULES OF THE LiYuTan RESERVOIR IN TAIWAN1

ABSTRACT: A procedure to apply genetic algorithm to optimize operation rules is proposed and applied to the LiYuTan Reservoir in Taiwan. The designed operation rules are operation zones with discount rates of water supply. The first step of the procedure is to predefine the shape of boundary curves of operation zones according to reservoir storage routing. Then, relatively fewer variables are used to describe the curves, and a last genetic algorithm (GA) is applied to optimize the curves. The procedure is applied to the newly built LiYuTan Reservoir for increasing domestic water demands. Shortage index is used to evaluate the performance of operation zones. A year is divided into 36 operational periods, with each month containing three operational periods. The shortage indexes calculated in operational periods are 9.81, 8.27, and 7.13, respectively, for the reservoir without operation rules, applying operation zones optimized by GA with encoding 36 storage levels for each curve, and adopting operation zones optimized by GA with encoding the curves with predefined shape. The average deficits for the three cases are 77.2, 43.6, and 33.3 (10⁴ m³/day), respectively. The results indicate that operation zones optimized by the proposed procedure have smaller shortage indexes and lower average deficits. In addition, the optimized operation zones have less variation and thus are more practical for operation. Conclusively, the proposed procedure utilizing GA to optimize operation zones with predefined shape can provide better and realistic outcomes through limited iterations.     

Optimal nonpoint source pollution control strategies for a reservoir watershed in Taiwan

The purpose of this study is to develop a model for optimal nonpoint source pollution control for the Fei-Tsui Reservoir watershed in Northern Taiwan. Several structural best management practices (BMPs) are selected to treat stormwater runoff. The complete model consists of two interacting components: an optimization model based on discrete differential dynamic programming (DDDP) and a zero-dimensional reservoir water quality model. A predefined procedure is used to locate suitable sites for construction of various selected BMPs in the watershed. In the optimization model, the objective function is to find the best combination of BMP type and placement, which minimizes the total construction and operation, maintenance, and repair (OMR) costs of the BMPs. The constraints are the water quality standards for total phosphorus (TP) and total suspended solids (TSS) concentrations in the reservoir. A zero-dimensional reservoir water quality model of the Vollenweider type is embedded in the optimization framework to simulate pollutant concentrations in Fei-Tsui Reservoir. The resulting optimal cost and benefit of water quality improvement are depicted by the model-derived trade-off curves. The modeling framework developed in the present study could be used as an efficient tool for planning a watershed-wide implementation of BMPs for mitigating stormwater pollution impact on the receiving water bodies.     

DECISION SUPPORT SYSTEM FOR RESERVOIR OPERATION1

ABSTRACT: A decision support system to determine reservoir releases in an uncertain environment during the dry season was developed. A key characteristic of the decision support system is its recursive procedure that processes observations to obtain the most feasible estimate. The system consists of three components: (1) a hydrologic model; (2) an optimization model, and (3) a fuzzy decision model. This methodology was applied to the operation of the Techi reservoir in central Taiwan. Three criteria (public water supply, irrigation, and hydropower) were taken into account within the operation process. Simulation results show that the decision support system can successfully assist government officials in determining operating policy for the Techi reservoir during the dry season. Also, the system is simple enough to lead to a rapid transfer of theoretical knowledge into practice.     

Methodology for Developing Flood Rule Curves Conditioned on El Niño‐Southern Oscillation Classification1

Lee, Se‐Yeun, Alan F. Hamlet, Carolyn J. Fitzgerald, and Stephen J. Burges, 2011. Methodology for Developing Flood Rule Curves Conditioned on El Niño‐Southern Oscillation Classification. Journal of the American Water Resources Association (JAWRA) 47(1):81‐92. DOI: 10.1111/j.1752‐1688.2010.00490.x Abstract: Regional climate varies on interannual and decadal time scales that in turn affect annual streamflows, flood risks, and reservoir storage deficits in mid‐summer. However, these variable elements of the climate system are generally not included in water resources operating policies that attempt to preserve a balance between flood risk and other water resources system objectives. A methodology for incorporating El Niño‐Southern Oscillation (ENSO) information in designing flood control curves is investigated. An optimization‐simulation procedure is used to develop a set of ENSO‐conditioned flood control rule curves that relate streamflow forecasts to flood control evacuation requirements. ENSO‐conditioned simulated flood risk and storage deficits under current operating policy are used to calibrate a unique objective function for each ENSO classification. Using a case study for the Columbia River Basin, we demonstrate that ENSO‐conditioned flood control curves constructed using the optimization‐simulation procedure consistently reduce storage deficits at a number of interrelated projects without increasing flood risk. For the Columbia Basin, the overall improvements in reservoir operations are relatively modest, and (in isolation) might not motivate a restructuring of flood control operations. However, the technique is widely applicable to a wide range of water resources systems and/or different climate indices.     

Multi-criteria analysis via the VIKOR method for prioritizing land-use restraint strategies in the Tseng-Wen reservoir watershed

It is important to adopt proper water and soil conservation and land-use planning in a watershed for lowering adverse impacts on reservoir water quality. Although reservoir watersheds occupy a large amount of land in Taiwan, high population density has exerted development pressures on such land. Therefore, the priority ranking of land-use restrictions for the subdivisions with different degrees of environmental vulnerability is necessary in watershed management. Since there are several criteria for evaluating the potential environmental impact from the subdivisions, multi-criteria analysis was applied as a technique for solving these problems in this study. The VIKOR method was applied to determine the best feasible solution according to the selected criteria, including geographical and meteorological factors. The objective of this study was to establish the priority ranking of land-use restrictions in the Tseng-Wen reservoir wastershed in southern Taiwan. The results show that subdivisions close to the outlet or reservoir area should have the priority of land-use restrictions.     

INDUSTRIAL WATER DEMAND WITH WATER REUSE1

ABSTRACT: This work establishes an industrial water demand (IWD) model for a short term estimate, which considers water reuse technologies and discharge regulations, for the integrated circuit (IC) industry in northern Taiwan. Based on the optimization of an industrial water cost system, a computerized system dynamics model (SD model) is developed to generate individual firm IWD using data from year 2000. A market IWD is further constructed for 25 1C firms in the study area and is approximated by an inverse logistic curve. Analytical results demonstrate that price elasticity varies with water price in cases involving water reuse.     

MULTIOBJECTIVE REAL‐TIME RESERVOIR OPERATION WITH A NETWORK FLOW ALGORITHM1

ABSTRACT: A network flow algorithm has been developed for the optimization of real‐time operation of a multiple reservoir system. Two purposes have been considered in the operation: flood control and hydropower generation. A special network structure was developed which allows the consideration of river routing. A multiobjective formulation is utilized thus allowing generation of a non‐dominated curve. The effect of imperfect forecast on the performance of the real‐time operation model is also evaluated. An application is made to a subsystem of the Brazilian hydroelectric system, located in the Paranapanema river basin. In this case study, the model showed good performance under the largest flood of the historical records.     

DERIVING THE NONLINEAR RISK-BENEFIT ALGORITHM FOR RESERVOIRS1

ABSTRACT: The Nonlinear Risk-Benefit (NRB) Algorithm includes risk as one of the objectives in a multiple-objective optimization problem. The NRB Algorithm is derived by extending the Surrogate Worth Trade-Off method to quadratic programming. This category of problem is common in water resources planning and design, especially multipurpose reservoir systems. Consequently, an example is given using the algorithm for optimally operating a multipurpose reservoir.     

COMPARATIVE STUDY OF OPTIMIZATION TECHNIQUES FOR IRRIGATION PROJECT PLANNING1

ABSTRACT: This study presents three optimization techniques for on‐farm irrigation scheduling in irrigation project planning: namely the genetic algorithm, simulated annealing and iterative improvement methods. The three techniques are applied to planning a 394.6 ha irrigation project in the town of Delta, Utah, for optimizing economic profits, simulating water demand, and estimating the crop area percentages with specific water supply and planted area constraints. The comparative optimization results for the 394.6 ha irrigated project from the genetic algorithm, simulated annealing, and iterative improvement methods are as follows: (1) the seasonal maximum net benefits are $113,826, $111,494, and $105,444 per season, respectively; and (2) the seasonal water demands are 3.03*10³ m³, 3.0*10³ m³, and 2.92*10³ m³ per season, respectively. This study also determined the most suitable four parameters of the genetic algorithm method for the Delta irrigated project to be: (1) the number of generations equals 800, (2) population size equals 50, (3) probability of crossover equals 0.6, and (4) probability of mutation equals 0.02. Meanwhile, the most suitable three parameters of simulated annealing method for the Delta irrigated project are: (1) initial temperature equals 1,000, (2) number of moves equal 90, and (3) cooling rate equals 0.95.     

A Simple Framework for Incorporating Seasonal Streamflow Forecasts into Existing Water Resource Management Practices1

Gong, Gavin, Lucien Wang, Laura Condon, Alastair Shearman, and Upmanu Lall, 2010. A Simple Framework for Incorporating Seasonal Streamflow Forecasts Into Existing Water Resource Management Practices. Journal of the American Water Resources Association (JAWRA) 46(3):574-585. DOI: 10.1111/j.1752-1688.2010.00435.x Abstract: Climate-based streamflow forecasting, coupled with an adaptive reservoir operation policy, can potentially improve decisions by water suppliers and watershed stakeholders. However, water suppliers are often wary of straying too far from their current management practices, and prefer forecasts that can be incorporated into existing system modeling tools. This paper presents a simple framework for utilizing streamflow forecasts that works within an existing management structure. Climate predictors are used to develop seasonal inflow forecasts. These are used to specify operating rules that connect to the probability of future (end of season) reservoir states, rather than to the current storage, as is done now. By considering both current storage and anticipated inflow, the likelihood of meeting management goals can be improved. The upper Delaware River Basin in the northeastern United States is used to demonstrate the basic idea. Physically plausible climate-based forecasts of March-April reservoir inflow are developed. Existing simulation tools and rule curves for the system are used to convert the inflow forecasts to reservoir level forecasts. Operating policies are revised during the forecast period to release less water during forecasts of low reservoir level. Hindcast simulations demonstrate reductions of 1.6% in the number of drought emergency days, which is a key performance measure. Forecasts with different levels of skill are examined to explore their utility.     

Urban water resources planning by using a modified particle swarm optimization algorithm

A new optimization algorithm by coupling the mutation process to the particle swarm optimization (PSO) is developed in this paper. This algorithm, entitled particle swarm optimization with mutation similarity (PSOMS), is successfully applied to an urban water resources management problem for the large city of Tabriz, Iran. The objective functions of the optimization problem are to minimize the cost, maximize water supply and minimize the environmental hazards. The constraints are physical limits such as pipelines capacity, ground water, the demand and the impact of conservation tools. Due to the parameters uncertainty, the water supply objective is modeled with fuzzy set theory and the objectives are then combined with compromise programming. The resulted single objective is solved using PSOMS, and its efficiency is then compared with the basic PSO and two kinds of genetic algorithms. Among them, PSOMS shows rapid convergence and suitable results compared to other methods. PSOMS is also improved to provide the Pareto frontier, which is needed to proper selecting of the optimal solutions in the uncertain conditions. Finally, the diversity of solutions is checked based on an indicator of the distances between different solutions, which show the efficiency of the PSOMS algorithm with respect to the genetic algorithm. Then by using the non-symmetric Kalai–Smorodinsky method a guideline is provided for comfort selection of the most preferred solution in the Pareto frontier. Based on these outcomes, the multi-objective PSOMS provides more appropriate results needed for urban systems management.     

ASSESSING FLOOD MITIGATION ALTERNATIVES IN SHIJR AREA IN METROPOLITAN TAIPEI1

The Keelung River Basin in northern Taiwan lies immediately upstream of the Taipei metropolitan area. The Shijr area is in the lower basin and is subject to frequent flooding. This work applies micromanagement and source control, including widely distributed infiltration and detention/ retention runoff retarding measures, in the Wudu watershed above Shijr. A method is also developed that combines a genetic algorithm and a rainfall runoff model to optimize the spatial distribution of runoff retarding facilities. Downstream of Wudu in the Shijr area, five dredging schemes are considered. If 10‐year flood flows cannot be confined in the channel, then a levee embankment that corresponds to the respective runoff retarding scheme will be required. The minimum total cost is considered in the rule to select from the regional flood mitigation alternatives. The results of this study reveal that runoff retarding facilities installed in the upper and middle parts of the watershed are most effective in reducing the flood peak. Moreover, as the cost of acquiring land for the levee embankment increases, installing runoff retarding measures in the upper portion of the watershed becomes more economical.     

DECISION SUPPORT FOR ALLOCATION OF WATERSHED POLLUTION LOAD USING GREY FUZZY MULTIOBJECTIVE PROGRAMMING1

ABSTRACT: This paper uses the grey fuzzy multiobjective programming to aid in decision making for the allocation of waste load in a river system under versatile uncertainties and risks. It differs from previous studies by considering a multicriteria objective function with combined grey and fuzzy messages under a cost benefit analysis framework. Such analysis technically integrates the prior information of water quality models, water quality standards, wastewater treatment costs, and potential benefits gained via in‐stream water quality improvement. While fuzzy sets are characterized based on semantic and cognitive vagueness in decision making, grey numbers can delineate measurement errors in data collection. By employing three distinct set theoretic fuzzy operators, the synergy of grey and fuzzy implications may smoothly characterize the prescribed management complexity. With the aid of genetic algorithm in the solution procedure, the modeling outputs contribute to the development of an effective waste load allocation and reduction scheme for tributaries in this subwatershed located in the lower Tseng‐Wen River Basin, South Taiwan. Research findings indicate that the inclusion of three fuzzy set theoretic operators in decision analysis may delineate different tradeoffs in decision making due to varying changes, transformations, and movements of waste load in association with land use pattern within the watershed.     

ANALYSIS OF FIRM WATER SUPPLY UNDER COMPLEX INSTITUTIONAL CONSTRAINTS1

River basin computer simulation studies often do not properly include the complex legal and institutional factors governing water allocation. These factors include formal water rights and informal borrowing agreements among the basin water users. An attempt has been made in this study to show that such factors can be included. We also show that an optimal, integrated approach to reservoir operations in a river basin can do much to alleviate the burden of new demands placed on available water resources. The procurement of a firm water supply for a proposed coal fired power plant is analyzed as a case study. An efficient river basin simulation model is used to determine the viability of a scheme for providing an annual firm water supply to the plant, with consideration of the existing water storage and demands within the basin. Given the hydrologic sequence considered, the model results show that the proposed strategy is viable in that the required firm water supply can be realized without causing harm to decreed water users in the basin. However, integrated diversion and reservoir operations are required to assure a desirable uniform rate of delivery of reusable effluent to the power plant.     

Automatic Calibration of Hydrologic Models With Multi‐Objective Evolutionary Algorithm and Pareto Optimization1

Abstract: In optimization problems with at least two conflicting objectives, a set of solutions rather than a unique one exists because of the trade‐offs between these objectives. A Pareto optimal solution set is achieved when a solution cannot be improved upon without degrading at least one of its objective criteria. This study investigated the application of multi‐objective evolutionary algorithm (MOEA) and Pareto ordering optimization in the automatic calibration of the Soil and Water Assessment Tool (SWAT), a process‐based, semi‐distributed, and continuous hydrologic model. The nondominated sorting genetic algorithm II (NSGA‐II), a fast and recent MOEA, and SWAT were called in FORTRAN from a parallel genetic algorithm library (PGAPACK) to determine the Pareto optimal set. A total of 139 parameter values were simultaneously and explicitly optimized in the calibration. The calibrated SWAT model simulated well the daily streamflow of the Calapooia watershed for a 3‐year period. The daily Nash‐Sutcliffe coefficients were 0.86 at calibration and 0.81 at validation. Automatic multi‐objective calibration of a complex watershed model was successfully implemented using Pareto ordering and MOEA. Future studies include simultaneous automatic calibration of water quality and quantity parameters and the application of Pareto optimization in decision and policy‐making problems related to conflicting objectives of economics and environmental quality.     

A STRATEGY FOR IMPLEMENTING BMPs FOR CONTROLLING NONPOINT SOURCE POLLUTION: THE CASE OF THE FEI‐TSUI RESERVOIR WATERSHED IN TAIWAN1

ABSTRACT: This paper describes a concerted effort by the Taiwan Water Resources Bureau, the City of Taipei, and the Bureau of Fei‐tsui Reservoir Management to protect the water quality in the Fei‐tsui Reservoir.The reservoir is the major source of water supply for over two million people in the metropolitan area of Taipei. Over the years the reservoir has suffered from siltation and more recently eutrophication. The sources of the pollution are traced to the hundreds of tea gardens, rice fields and other agricultural areas in the watershed and to urban sources such as construction sites. Large amounts of nutrients enter the reservoir by way of storm water runoff during storm or typhoon events. Since 1999, various agencies have worked to initiate an effort to reduce nonpoint pollution in the Fei‐Tsui Reservoir watershed. Practices being considered include nonstructural measures such as nutrient management, and structural measures such as swales, detention basins, and wetlands, in addition to erosion and sediment control methods. A number of field tests have been completed on the performance of selected best management practices (BMPs). A strategy for implementing the BMPs at the watershed scale has been developed based on a total maximum daily load (TMDL) analysis that is reported in this paper.     

Autonomous real-time adaptive management of soil salinity using a receding horizon control algorithm: a pilot-scale demonstration

Soil salinization is a potentially negative side effect of irrigation with reclaimed water. While optimization schemes have been applied to soil salinity control, these have typically failed to take advantage of real-time sensor feedback. This study incorporates current soil observation technologies into the optimal feedback-control scheme known as Receding Horizon Control (RHC) to enable successful autonomous control of soil salinization. RHC uses real-time sensor measurements, physically-based state prediction models, and optimization algorithms to drive field conditions to a desired environmental state by manipulating application rate or irrigation duration/frequency. A simulation model including the Richards equation coupled to energy and solute transport equations is employed as a state estimator. Vertical multi-sensor arrays installed in the soil provide initial conditions and continuous feedback to the control scheme. An optimization algorithm determines the optimal irrigation rate or frequency subject to imposed constraints protective of soil salinization. A small-scale field test demonstrates that the RHC scheme is capable of autonomously maintaining specified salt levels at a prescribed soil depth. This finding suggests that, given an adequately structured and trained simulation model, sensor networks, and optimization algorithms can be integrated using RHC to autonomously achieve water reuse and agricultural objectives while managing soil salinization.     

A CODED ALGORITHM FOR CAPACITY EXPANSION OF A WATER QUALITY MANAGEMENT SYSTEM1

ABSTRACT: This paper describes a mathematical model, an algorithm and a computer program that were specially developed to study the problem of a water quality management system undergoing a rapidly increasing environmental stress. The model output will determine the locations, sizes and the timing of construction of new treatment plants plus an overall treatment plant operating policy so that environmental standards are maintained at a minimum cost. The model, as formulated, is a 0-1 mixed integer programming problem which is solved by decomposing it into a capital budgeting problem (solved by Little's branch and bound algorithm) and an operational policy problem (solved by linear programming). The coded algorithm (in FORTRAN 10) has been tested with a semi-realistic example.     

APPLICATION OF CROP YIELD FUNCTIONS IN RESERVOIR OPERATION1

ABSTRACT: A model is developed for real-time operation of an irrigation reservoir with the objective of maximizing the value of multiple crop yields during a growing season. The model employs monthly additive and product forms of crop yield functions for dry matter and grain crops, respectively. The resulting nonlinear optimization model uses a log transform to reduce nonlinearities in the model. An application of the proposed model is compared to a common operating rule used in simulation models. The proposed model results were better in terms of net benefits from crop yields. The model uses GAMS (General Algebraic Modeling System) language. It requires an IBM-compatible microcomputer and is suitable for use by a reservoir manager.