A COMPARISON OF SOME DYNAMIC,LINEAR AND POLICY ITERATION METHODS FOR RESERVOIR OPERATION1

Within the past few years, a number of papers have been published in which stochastic mathematical programming models, incorporating first order Markov chains, have been used to derive alternative sequential operating policies for a multiple purpose reservoir. This paper attempts to review and compare three such mathematical modeling and solution techniques, namely dynamic programming, policy iteration, and linear programming. It is assumed that the flows into the reservoir are serially correlated stochastic quantities. The design parameters are assumed fixed, i.e., the reservoir capacity and the storage and release targets, if any, are predetermined. The models are discrete since the continuous variables of time, volume, and flow are approximated by discrete units. The problem is to derive an optimal operating policy. Such a policy defines the reservoir release as a function of the current storage volume and inflow. The form of the solution and some of the advantages, limitations and computational efficiencies of each of the models and their algorithms are compared using a simplified numerical example.     

STOCHASTIC DYNAMIC PROGRAMMING FOR OPTIMUM RESERVOIR OPERATION1

ABSTRACT. For a multipurpose single reservoir a deterministic optimal operating policy can be readily devised by the dynamic programming method. However, this method can only be applied to sets of deterministic stream flows as might be used repetitively in a Monte Carlo study or possibly in a historical study. This paper reports a study in which an optimal operating policy for a multipurpose reservoir was determined, where the optimal operating policy is stated in terms of the state of the reservoir indicated by the storage volume and the river flow in the preceding month and uses a stochastic dynamic programming approach. Such a policy could be implemented in real time operation on a monthly basis or it could be used in a design study. As contrasted with deterministic dynamic programming, this method avoids the artificiality of using a single set of stream flows. The data for this study are the conditional probabilities of the stream flow in successive months, the physical features of the reservoir in question, and the return functions and constraints under which the system operates.     

A COMPARISON OF TWO METHODS FOR MULTIOBJECTIVE OPTIMIZATION FOR RESERVOIR OPERATION1

ABSTRACT: Two major objectives in operating the multireservoir system of the Upper Colorado River basin are maximization of hydroelectric power production and maximization of the reliability of annual water supply. These two objectives conflict. Optimal operation of the reservoir system to achieve both is unattainable. This paper seeks the best compromise solution for an aggregated reservoir as a surrogate of the multireservoir system by using two methods: the constraint method and the method of combined stochastic and deterministic modeling. Both methods are used to derive the stationary optimal operating policy for the aggregated reservoir by using stochastic dynamic programming but with different objective functions and minimum monthly release constraints. The resulting operating policies are then used in simulated operation of the reservoir with historical inflow records to evaluate their relative effectiveness. The results show that the policy obtained from the combination method would yield more hydropower production and higher reliability of annual water supply than that from the constraint-method policy.     

SEASONAL ARIMA INFLOW MODELS FOR RESERVOIR SIZING1

ABSTRACT: The reliable sizing of reservoirs is a very important task of hydraulic engineering. Although many reservoirs throughout the world have been designed using Rippl's mass curves with historical inflow volumes at the dam site, this technique is now considered outdated. In this paper, synthetic series of monthly inflows are used as an alternative to historical inflow records. These synthetic series are generated from stochastic SARIMA (Seasonal Autoregressive Integrated Moving Average) models. The analyzed data refer to the planned Almopeos Reservoir on the Almopeos River in Northern Greece with 19‐year monthly inflow series. The analysis of this study demonstrates the ability of SARIMA models, in conjunction with the adequate transformation, to forecast monthly inflows of one or more months ahead and generate synthetic series of monthly inflows that preserve the key statistics of the historical monthly inflows and their persistence Hurst coefficient K. The forecasted monthly inflows would be of help in evaluating the optimal real time reservoir operation policies and the generated synthetic series of monthly inflows can be used to provide a probabilistic framework for reservoir design and to cope with the situation where the design horizon of interest exceeds the length of the historical inflow record.     

THE EFFECTS OF RISK AND RELIABILITY ON OPTIMAL RESERVOIR DESIGN1

A chance-constrained linear programming model, which utilizes multiple linear decision rules and is useful for river basin planning, is used to evaluate the effects of risk and reliability on optimal reservoir design. Streamflow forecasts or predictions can be explicitly included in the linear program. The risk associated with the predictions is included in the model through the use of cumulative distribution functions (CDF) of streamflows which are conditioned on the predictions. A multiple-purpose reservoir on the Gunpowder River in Maryland is used to illustrate the effectiveness of the model. In order to provide the decision makers with complete and useful information, trade-off curves relating minimum reservoir capacity (a surrogate for dam costs), water supply and flood control targets, and the reliability of achieving the targets are developed. The trade-off curves may enhance the decision maker's ability to select the best dam capacity, considering technological and financial constraints as well as the trade-offs between targets, risks, and costs.     

OPTIMIZATION OF MULTIPURPOSE RESERVOIR SYSTEM OPERATION1

ABSTRACT: The goal programming approach for multipurpose reservoir operation has been proposed and applied to the Bhadra reservoir system, having irrigation and hydropower production as dual purposes, in India. The objective of the model is to satisfy sequentially a series of operating criteria. Two goal programming models, one with the objective function as minimizing the deviations from storage targets and the other with the objective function as minimizing the deviations from release targets, have been formulated and applied to the reservoir system under study. The results proved that the model with release targets is preferred over the model with storage targets for determining operational policies for multipurpose reservoir system.     

OPTIMAL RESERVOIR MANAGEMENT AND CROP PLANNING UNDER DETERMINISTIC AND STOCHASTIC INFLOWS1

ABSTRACT: This study analyzes planning under deterministic and stochastic inflows for the Mayurakshi project in India. Models are developed to indicate the optimal storage of reservoir water, the transfer of water to the producing regions, and the spillage of water from the reservoir, if needed. A deterministic programming model was first formulated to represent the existing situation. A chance-constrained model then was constructed to evaluate potential violations of the deterministic model. Both models were quantified for the command area. Data were collected from surveys of the area and from government agencies. Both the deterministic and change-constrained models suggest a more intensive cropping program in the region. Both emphasize more dependence on rabi and less on kharif crops. The chance-constrained especially suggests use of more water in the rabi season. Important chances in cropping programs and labor use take place under the chance-constrained model.     

RESERVOIR RELEASES TO USES WITH DIFFERENT RELIABILITY REQUIREMENTS1

ABSTRACT: Releases from a reservoir may be allocated to a number of uses, each of which may require a given volume of water at a different reliability. The paper provides a method that can be used to estimate the volume of water associated with a given reliability for each use of water when the proportion of releases allocated to each use is known. These results can be used to evaluate the meeting of specified objectives under a published release policy derived by stationary stochastic dynamic programming. The results can also be used to solve water allocation problems when the probability distribution of available water is known (or can be estimated) and water has multiple uses, each of which has different volume and reliability requirements.     

Inexact fuzzy-stochastic mixed-integer programming approach for long-term planning of waste management – Part A: Methodology

In this study, an inexact fuzzy chance-constrained two-stage mixed-integer linear programming (IFCTIP) approach is proposed for supporting long-term planning of waste-management systems under multiple uncertainties in the City of Regina, Canada. The method improves upon the existing inexact two-stage programming and mixed-integer linear programming techniques by incorporating uncertainties expressed as multiple uncertainties of intervals and dual probability distributions within a general optimization framework. The developed method can provide an effective linkage between the predefined environmental policies and the associated economic implications. Four special characteristics of the proposed method make it unique compared with other optimization techniques that deal with uncertainties. Firstly, it provides a linkage to predefined policies that have to be respected when a modeling effort is undertaken; secondly, it is useful for tackling uncertainties presented as intervals, probabilities, fuzzy sets and their incorporation; thirdly, it facilitates dynamic analysis for decisions of facility-expansion planning and waste-flow allocation within a multi-facility, multi-period, multi-level, and multi-option context; fourthly, the penalties are exercised with recourse against any infeasibility, which permits in-depth analyses of various policy scenarios that are associated with different levels of economic consequences when the promised solid waste-generation rates are violated. In a companion paper, the developed method is applied to a real case for the long-term planning of waste management in the City of Regina, Canada.     

A SEPARABLE LINEAR ALGORITHM FOR HYDROPOWER OPTIMIZATION1

ABSTRACT: A deterministic, separable, linear algorithm is presented for maximizing aggregate hydropower production. The method is iterative and amenable to solution using standard LP software. The utility of the technique is demonstrated using several test applications involving a hypothetical single-purpose hydropower reservoir and a monthly increment 20-year flow record from the Gunpowder River in Maryland. The separable linearized forms solved quickly using MPSX on a variety of IBM hardware: 3090-400 VF, 3084 QX, dual processor 4381-3, and an AT/370 personal computer. For comparison purposes, the original nonlinear nonseparable version of the model was also solved using MINOS. This yielded a value of aggregate hydropower marginally higher than that using MPSX. The separable, linearized methodology proved to be a useful and an efficient means of generating good starting points for MINOS. The use of these warm starts effected substantial reductions in MINOS execution times.     

OPTIMAL CONTROL OF RESERVOIR RELEASES TO MINIMIZE SEDIMENTATION IN RIVERS AND RESERVOIRS1

ABSTRACT: An optimal control methodology and computational model are developed to evaluate multi‐reservoir release schedules that minimize sediment scour and deposition in rivers and reservoirs. The sedimentation problem is formulated within a discrete‐time optimal control framework in which reservoir releases represent control variables and reservoir bed elevations, storage levels, and river bed elevations represent state variables. Constraints imposed on reservoir storage levels and releases are accommodated using a penalty function method. The optimal control model consists of two interfaced components: a one‐dimensional finite‐difference simulation module used to evaluate flow hydraulics and sediment transport dynamics, and a successive approximation linear quadratic regulator (SALQR) optimization algorithm used to update reservoir release policies and solve the augmented control problem. Hypothetical two‐reservoir and five‐reservoir networks are used to demonstrate the methodology and its capabilities, which is a vital phase towards the development of a more robust optimal control model and application to an existing multiple‐reservoir river network.     

THE OPTIMALITY OF CAPACITY SHARING IN STOCHASTIC DYNAMIC PROGRAMMING PROBLEMS OF SHARED RESERVOIR OPERATION1

ABSTRACT: This paper is concerned with finding an optimal allocation of water entitlements for each of two users of water who share a reservoir. Two instruments of allocation are considered. The first, release sharing, involves sharing the releases from the reservoir; the second, capacity sharing, is concerned with allocating to each user of water a share of inflows, reservoir capacity and leakage and evaporation losses. Stochastic dynamic programming problems of reservoir operation under each type of sharing arrangement are formulated. It is shown that the maximum discounted expected profit from reservoir operation over the life of the storage using capacity sharing is at least as large as that obtained using release sharing and that release sharing is not Pareto efficient.     

ECONOMIC ANALYSIS OF IRRIGATION PRODUCTION FUNCTIONS: AN APPLICATION OF LINEAR PROGRAMMING1

ABSTRACT: Economic analysis of irrigation production functions is discussed using linear programming. The method provides advantages over the partial one-crop type of analyses because it captures intercrop tradeoffs in water and land use in response to economic policies or changes in water supply. A numerical example is used.     

OPTIMAL CAPACITIES OF WATER SUPPLY RESERVOIRS IN SERIES AND PARALLEL1

ABSTRACT: The planning of water supply reservoirs has traditionally been based on the Rippl or sequent peak analysis which applies to the design of a single reservoir. This paper incorporates the sequent peak method as the central feature in establishing a procedure for determining the sizes of several potential reservoirs located in a system of one or more rivers. Separate algorithms are developed for sites on parallel streams and for sites on the same stream. In both cases the approach is to find the combination of reservoirs which can satisfy a given constant monthly demand at a minimum total construction cost. It is shown that both problems can be cast in the form of a dynamic programming problem. A more complex system is then a combination of reservoirs in parallel and in series. An extension is given if the monthly demand is not constant but each reservoir satisfies a constant fraction of the monthly demand.     

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.     

A STOCHASTIC DYNAMIC PROGRAMMING BASED APPROACH TO THE OPERATION OF A MULTI-RESERVOIR SYSTEM1

ABSTRACT: An heuristic iterative technique based upon stochastic dynamic programming is presented for the analysis of the operation of a three reservoir ‘Y’ shaped hydroelectric system. The technique is initiated using historical inflow data for the downstream reservoir. At each iteration the optimal policies for the downstream hydroelectric generating unit are used to provide relative weightings or targets for operation of upstream reservoirs. New input inflows to the downstream reservoir are then obtained by running the historical streamflow record through the optimal policies for the upstream reservoirs. These flows are then used to develop a new operating policy for the downstream reservoir and hence new targets for the upstream reservoirs. The process is continued until the operating policies for each reservoir provide the same overall system benefit for two successive iterations. Results obtained from the procedure are compared to the results obtained by historical operation of the system. The procedure is shown to develop operating policies which give benefits which are as close to the historical benefits as can be expected given the choice of the number of storage state variables.     

AN APPLICATION OF MATHEMATICAL PROGRAMMING IN PLANNING SURFACE WATER STORAGE1

Designing a surface reservoir involves the concept of reservoir yield. This concept embodies three basic information items: hydrologic regime, active storage volume, and reservoir release policy. In the actual case presented below, the magnitude of the active storage was prescribed by a legal procedure, so that the planning issue became that of determining the reservoir yield given the hydrological information. A stochastic dynamic programming model was formulated to derive a schedule of seasonal optimal reservoir releases and their respective probabilities of occurrence. This schedule is the reservoir yield. The yearly cycle was divided into three seasons representing the actual climatic conditions, and conditional probabilities linking streamflows in consecutive seasons were estimated. An operating policy was postulated, based on the same set of legal decisions that prescribed the active storage volume, and target reservoir releases were assumed. Similarly, target storages at the end of each season were set up. The optimizing/ minimizing criterion in the dynamic programming formulation was the sum of squares of deviations of actual releases and final storage volumes from their respective targets.     

MULTI-OBJECTIVE RESERVOIR OPERATION INCLUDING TURBIDITY CONTROL1

ABSTRACT: The operational problems of a reservoir are expressed by three coordinates: space, time stage, and objective. The operational procedure is formulated using dynamic programming as a multi-objective problem. After comparing the scalar and the vector optimization, the scalar optimization technique is applied to turbidity analysis in a reservoir.     

MULTIOBJECTWE OPTIMIZATION OF RESERVOIR SYSTEMS OPERATION1

ABSTRACT: Development of optimal operational policies for large-scale reservoir systems is often complicated by a multiplicity of conflicting project uses and purposes. A wide range of multiobjective optimization methods are available for appraising tradeoffs between conificting objectives. The purpose of this study is to provide guidance as to those methods which are best suited to dealing with the challenging large-scale, nonlinear, dynamic, and stochastic characteristics of multireservoir system operations. As a case study, the selected methodologies are applied to the Han River Reservoir System in Korea for four principal project objectives: water supply and low flow augmentation; annual hydropower production, reliable energy generation, and minimization of risk of violating firm water supply requirements. Additional objectives such as flood control are also considered, but are imposed as fixed constraints.     

AN ARCHETYPAL LINEAR PROGRAMMING MODEL FOR OPTIMAL INTERTEMPORAL ALLOCATION OF GROUNDWATR TO AGRICULTURAL ACTIVITIES1

ABSTRACT: In the past, researchers have applied a variety of analytical techniques for maximizing the present value of a stock resource - simulation, calculus of variations, stochastic dynamic programming, and optimal control theory. This paper presents a more operational approach - linear programming. A simplified, broadly drawn example from Southwest agriculture is used for demonstrating the model's structure and output. The method is based on a set of state transformation operations that prevent the additivity assumption of linear programming from being violated.