Reservoir Operation Model Research Articles

Neural Network-Based Simulation-Optimization Model for Reservoir Operation

There have been several attempts to use combined simulation-optimization models to solve reservoir operation problems efficiently. In many cases, complex simulation models are available, but direct incorporation of them into an optimization framework is computationally prohibitive. To overcome this problem, in this study, a backpropagation neural network is trained to approximate the simulation model developed for the Chennai city water supply problem. The neural network is used as a submodel in a Hooke and Jeeves nonlinear programming model to find “near optimal policies.” The results are further refined using the conventional simulation-optimization model.

A short-term reservoir operation model for multicrop irrigation

An integrated model is developed for short-term yearly reservoir operation for irrigation of multiple crops. The model optimizes a measure of annual crop production, starting from the current period in real time. Reservoir storage at the beginning of a period, inflow during the previous period, crop soil moisture values and crop production already achieved up to the beginning of the period are used as inputs to the model. The solution specifies the reservoir release and optimal irrigation allocations to individual crops during an intra-seasonal period. The model overcomes some of the limitations of an earlier model developed by Mujumdar & Ramesh (1997) by replacing the two dynamic programming (DP) formulations with a single linear programming (LP) formulation. Application of the model is studied through a case study in India.

Bi-objective dynamic programming for trading off hydropower and irrigation

Reservoir management is examined from the bicriterion viewpoints of hydropower generation and irrigation. Specifically, a decision support system is developed which includes constructing the set of non-inferior solutions, using the constraint, penalty and weighting methods. The reservoir operation model, as well as the optimization and simulation software, were initially provided by the R & D Division of the French Electrical Utility (Electricité de France); the method is based on a Stochastic Dynamic Programming—SDP—model with daily time steps, using 50 annual values of daily historical inflows and marginal electricity production costs. The water requirement for irrigation is taken either as preset using a demand curve, or as a soil reservoir model based on SDP and running simultaneously with the reservoir model. The model yields trade-off curves of electrical production losses versus, respectively, irrigated surface area, crop yield, and agricultural benefits. A numerical application illustrates the methodology in the French case of the Bort reservoir, in the Dordogne valley. The model reveals that releasing water for irrigation could present economic advantages within certain constraints on maximum irrigated surface area and minimum water price. Consequently, the current economic justification of single purpose reservoir management for hydropower ought to be changed. Negotiation should be initiated for the purpose of selecting one of the non-dominated reservoir management policies, hereby minimizing the distance of the “most preferred” solution point from the ideal one or by using the Nash cooperative game solution. Multicriterion decision-making seems to be a promising operational management tool for hydrosystems.

Incorporating inflow uncertainty into risk assessment for reservoir operation

An attempt of using stochastic hydrologic technique to assess the intrinsic risk of reservoir operation is made in this study. A stochastic simulation model for reservoir operation is developed. The model consists of three components: synthetic generation model for streamflow and sediment sequences, one-dimensional delta deposit model for sediment transport processes in reservoirs, and simulation model for reservoir operation. This kind of integrated simulation model can be used to simulate not only the inflow uncertainty of streamflow and sedimentation, but also the variation in operation rules of reservoirs. It is herein used for the risk assessment of a reservoir, and the simulation is performed for different operation scenarios. Simulation for the 100-year period of sediment transport and deposition in the river-reservoir system indicates that the navigation risk is much higher than that of hydropower generation or sediment deposition in the reservoir. The risk of sediment deposition at the river-section near the backwater profile is also high thereby the navigation at the river-segment near this profile takes high risk because of inadequate navigation depth.

Real‐time reservoir operation for irrigation

A short‐term real‐time reservoir operation model is developed for irrigation of multiple crops. Two major features of the model distinguish it from the earlier models which address the problem of reservoir operation for irrigation: its capability to consider interdependence of crop water allocations across time periods and its ability to provide an adaptive release policy. Integration of reservoir operation with on‐farm utilization of water by crops is achieved through the use of two models: An operating policy model, which optimizes reservoir releases across time periods in a year, and an allocation model, which optimizes irrigation allocations across crops within a time period. The model is applied to a case study of an existing reservoir in India.

Sensitivity of reservoir operation performance to climatic change

The potential impacts of changing climatic conditions on the operational performance of water resource systems was investigated in this paper. A multi-site streamflow generation model was used to synthesize potential monthly flow sequences reflecting two different sets of climatic conditions. The generated data were subsequently employed as input to a reservoir operation model that was used to determine the reservoir response to the inflow resulting from the implementation of the reservoir operating policy. The performance of an example reservoir system, the Shellmouth Reservoir located in the Canadian province of Manitoba, was evaluated and compared for the two sets of conditions. The operational performance was evaluated in terms of the reliability of the system for meeting the three purposes of the actual reservoir. The reservoir performance was determined to be sensitive to the inflow data. The results indicate that climatic change has potentially important implications for the operation of the example reservoir system.

An Integrated Model for Optimal Reservoir Operation for Irrigation of Multiple Crops

An integrated model is developed, based on seasonal inputs of reservoir inflow and rainfall in the irrigated area, to determine the optimal reservoir release policies and irrigation allocations to multiple crops. The model is conceptually made up of two modules. Module 1 is an intraseasonal allocation model to maximize the sum of relative yields of all crops, for a given state of the system, using linear programming (LP). The module takes into account reservoir storage continuity, soil moisture balance, and crop root growth with time. Module 2 is a seasonal allocation model to derive the steady state reservoir operating policy using stochastic dynamic programming (SDP). Reservoir storage, seasonal inflow, and seasonal rainfall are the state variables in the SDP. The objective in SDP is to maximize the expected sum of relative yields of all crops in a year. The results of module 1 and the transition probabilities of seasonal inflow and rainfall form the input for module 2. The use of seasonal inputs coupled with the LP‐SDP solution strategy in the present formulation facilitates in relaxing the limitations of an earlier study, while affecting additional improvements. The model is applied to an existing reservoir in Karnataka State, India.

Proposed Daily Streamflow-Forecasting Model for Reservoir Operation

Daily reservoir operations rely increasingly on accurate streamflow forecasts, which today are derived from computer modeling. This paper documents development of a hydrologic-process-based daily streamflow-forecasting model with a built-in parameter-estimation routine. The model [called the National Taiwan University (NTU) Daily Model] closely captures the physical reality while maintaining the robustness of the statistical model, both of which must be considered in the day-to-day operations of a reservoir. Several equations are used to describe various simplified hydrologic processes, such as direct runoff, baseflow, infiltration, evapotranspiration, and change in ground-water storage. It is assumed that these equations can generate total daily streamflow from precipitation and initial conditions. The model parameters are automatically determined by a built-in Gauss-Newton algorithm. Model performance was analyzed by five statistical error measurements, including correlation coefficient, coefficient of efficiency, relative error, standard error, and volume error. The proposed model was calibrated and verified using historical streamflow data from three gauging stations in Taiwan. The results obtained indicate that the NTU Daily Model for streamflow forecasting is satisfactory for use in daily operation of the reservoir.

A multiobjective reservoir operation model with stochatic inflows

This study is concerned with the operation and control of multipurpose reservoir systems. The problem involves finding appropriate release decisions from various reservoirs in the system in order to satisfy multiple objectives which may be conflicting. These objectives include municipal and industrial (M&I) water supply, downstream water supply, flood control, hydroelectric power generation, and recreation, among others. This paper presents a mathematical model which reflects the three important characteristics of the problem which are: multiple objectives, stochastic inflows, and large-scale nature of the system. The model is based on the chance-constrained goal programming (CCGP) concept. The application of the model to a three-reservoir system in Oklahoma is also presented in this paper.

DDC·FDC Rule Curves for Large Reservoirs in Tropical Monsoon Areas

A set of DDC·FDC Rule Curves for Bhumibol and Sirkit Reservoirs in Thailand weredeveloped. The concept is applicable for operation of large reservoirs in any tropical monsoon region with rice paddy agriculture. They assume that the irrigation demand for a reservoir is reciprocal to inflow to the reservoir, which was remarkably well satisfied in the Bhumibol Reservoir in Chao Phraya River Basin. DDC·FDC Rule Curves are so named since they are the solutions of a chance constraint reservoir operation model where drought duration curves (DDC) and flood duration curves (FDC) were used as deterministic equivallence of sum of future streamflows. The obtained rule curves suggested that the Chao Phraya reservoir system was not large enough to provide as much water required now. In order to hold a 5% failture rate, with the storage of 1990 and 1991, as much as 60% of supply cutmay be necessary.

Risk-based parameter selection for short-term reservoir operation

A methodology has been developed to determine the optimal operating horizon for short-term reservoir operation. The modelling procedure includes: (1) an adaptive forecasting algorithm; (2) a real-time reservoir operation model; (3) a multiobjective compromise programming algorithm. The compromise programming algorithm utilizes a multiobjective compromise between the conflicting objectives of hydrologic forecast reliability and reservoir operation penalties. The algorithm requires predetermined weights defining the relative importance of the two objectives. This paper presents a risk-based methodology developed to help a decision maker in selecting the appropriate weights. Reliability, vulnerability, and resiliency are used as three risk-based criteria for assisting in selection of the weights. The methodology is presented in the form of a rule-based system designed to aid the user in the selection process. The methodology is illustrated through an application of the technique to the operation of Green Reservoir in Kentucky. The reservoir model is implemented for four different historical periods and its performance in terms of total penalties, reliability, vulnerability, and resiliency is presented as a function of the multi-objective weighting parameters. The rule-based expert system is then used for deriving recommendations regarding the selection of appropriate weights. The paper also discusses the rule-based expert system development tool, used in this research.

Optimization of short-term operation of a single multipurpose reservoir — a goal programming approach

The complex real-time reservoir operation model consists of flow forecasting, reservoir operations planning, and multiobjective modules. In the present paper the original reservoir operations planning module based on a linear programming (LP) formulation is substituted with a goal programming (GP) formulation. The advantage of applying goal programming instead of linear programming lies in the smaller input data requirement, and simpler model formulation. The LP approach requires two-dimensional information while the GP approach needs only one-dimensional information. The impact of change of the operations planning routine on the results of the complex model is observed and commented on a case study. The application of GP instead of LP is found viable, in certain cases even advantageous, giving a better overall objective function value. Key words: reservoir operation, optimization.

Risk-based performance criteria for real-time reservoir operation

Reliability, resiliency, and vulnerability criteria are formulated as risk-based performance measures for the evaluation of a real-time reservoir operation model. The reservoir operation model includes a multi-objective compromise programming algorithm to select, in real time, an optimal operating horizon for the reservoir operation. The utility of the risk-based performance criteria for comparing operational strategies resulting from the selection of different parameters for the compromise programming algorithm is demonstrated. Although trade-offs exist between the performance evaluators, it is shown that appropriate compromises can be reached between the conflicting modeling goals. Key words: reservoir operation, real time, risk, optimization.

An improved methodology for short‐term operation of a single multipurpose reservoir

The existence of several problems associated with the short‐term operation of multipurpose reservoirs has been demonstrated in the past. A principal difficulty arises from uncertainties inherent in the prediction of future stream flows which are required to effect the requisite reservoir operation. It has been shown that the performance of the reservoir operation, as determined by an optimization model, is a function of the operating horizon. This results from the existence of a relationship between the operating horizon and both the reliability of the inflow forecasts and the reservoir operation. With an increase in the length of the operating horizon, the reliability of the forecasts will necessarily decrease. However, the reservoir operation model yields an improved operation on the assumption that the forecasts are accurate. An improved methodology is presented which considers the operating horizon as a decision variable which can change in real time. The optimal value of the operating horizon is selected based on the trade‐off between more reliable inflow forecast for shorter horizons and better reservoir operation associated with the use of longer operating horizons. The benefits of the new methodology are illustrated through an application of the technique in simulating the operation of Green Reservoir in Kentucky.

Including daily constraints in a monthly reservoir operation model for low-flow

In many real-world situations, operation of water resources systems are subject to constraints which are formulated on a daily basis. Since mathematical models (simulation or optimization models) are often developed on a monthly basis to avoid dimensionality problems and to reduce necessary computer time, some degree of approximation is necessary. Two examples are presented in order to show how this approximation can be done: • - a reservoir is operated for hydroelectric power production and low-flow augmentation is provided on a daily basis as a function of inflows (Lech River System in Germany) • - a reservoir is operated optimally for daily low-flow augmentation at a control gage downstream. The model includes also mandatory releases as constraints (Wupper River System in Germany). In the first example the daily requirements are part of the constraints in the system, while in the second example the daily constraint concerns directly the objective function. Further, in the first case water can be saved and in the second case more water is needed as compared to calculations on a monthly basis. These examples are presented to show two solutions using envelope curves but other possibilities (i.e., regression analysis, constraints, coefficients) could be considered.

Improved Hourly Reservoir Operation Model

The modifications described in the following greatly facilitate the attainment of a good initial feasible policy and eliminate the need for the LP, minimize multiple start-sups and shutdowns of the generating units, and result in better Optimal solutions, although global optima cannot be guaranteed.

A Chance Constrained Optimization Model for reservoir design and operation

A chance constrained linear programing model which employs multiple linear decision rules is developed. The model incorporates explicitly the stochastic nature of the streamflow process, can be used in design and/or management situations, does not significantly restrict the operating policy prior to solution, and is economically and computationally feasible. A portion of the Yakima River system is modeled to demonstrate the use of the multiple linear decision rules.

Linear decision rule in reservoir management and design: 3. Direct capacity determination and intraseasonal constraints

A chance‐constrained model for multipurpose reservoir operation is reviewed and extended. The model minimizes required capacity subject to chance constraints for water supply, recreation, and flood control. A linear decision rule is used to make commitments based on existing storage. A study of the structure of this linear programing formulation reveals a direct solution in the special case when as much water as possible is reliably committed, a reasonable case for water supply. The structure of the model requires an adequate degree of continuity. This requirement generally restricts the usefulness of the model to multipurpose reservoirs. Using an execution rule to specify when within a decision period water will be released, we can extend the model to include constraints on storage within a decision period. Increasing the length of a decision period is seen to produce a large increase in reservoir capacity due to the making of commitments for a longer future of uncertain inflows. Numerical examples are provided that demonstrate the increased capacity required in a seasonal model over that required in a monthly model and the effect of using intraseasonal constraints.