Development of Operational Policy for a Multi-reservoir System in India using Genetic Algorithm

A genetic algorithm (GA) and a backward moving stochastic dynamic programming (SDP) model has been developed for derivation of operational policies for a multi-reservoir system in Kodaiyar River Basin, Tamil Nadu, India. The model was developed with the objective of minimizing the annual sum of squared deviation of desired target releases. The total number of population, crossover probability and number of generations of the GA model was optimized using sensitivity analysis, and penalty function method was used to handle the constraints. The policies developed using the SDP model was evaluated using a simulation model with longer length of inflow data generated using monthly time stepped Thomas–Fiering model. The performance of the developed policies were evaluated using the performance criteria namely, the monthly frequency of irrigation deficit (MFID), Monthly average irrigation deficit (MAID), Percentage monthly irrigation deficit (PMID), Annual frequency of irrigation deficit (AFID), Annual average irrigation deficit (AAID), and Percentage annual irrigation deficit (PAID). Based on the performance, it was concluded that the robostic, probabilistic, random search GA resulted in better optimal operating policies for a multi-reservoir system than the SDP models.     

Genetic Algorithm for Optimal Operating Policy of a Multipurpose Reservoir

This paper presents a Genetic Algorithm (GA) model for finding the optimal operating policy of a multi-purpose reservoir, located on the river Pagladia, a major tributary of the river Brahmaputra. A synthetic monthly streamflow series of 100 years is used for deriving the operating policy. The policies derived by the GA model are compared with that of the stochastic dynamic programming (SDP) model on the basis of their performance in reservoir simulation for 20 years of historic monthly streamflow. The simulated result shows that GA-derived policies are promising and competitive and can be effectively used for reservoir operation.     

Effect of inflow forecast accuracy and operating time horizon in optimizing irrigation releases

In this study, application of Genetic Algorithms (GA) is demonstrated to optimize reservoir release policies to meet irrigation demand and storage requirements. As it is commonly recognized that accuracy of inflow forecast and operating time horizon affects the optimal policies, a trial-and-error approach is suggested to identify the appropriate trade-off between forecast accuracy and operating horizon. The flexibility offered by GA to set up and evaluate objective functions is exploited towards this end. The results are also compared with Linear Programming (LP) model. It is concluded that forecasts models of high accuracy are desirable, particularly when the system is to be operated for periods of high demand. In such cases, the optimization with longer time horizon ensures achievement of the objective more uniformly over the period of operation. The performance of GA is found to be better than LP, when forecast model of higher accuracy and longer period of operating horizon are considered for optimization.     

Synchronizing Artificial Intelligence Models for Operating the Dam and Reservoir System

It is remarkable that several hydrological parameters have a significant effect on the reservoir operation. Therefore, operating the reservoir system is complex issue due to existing the nonlinearity hydrological variables. Hence, determining modern model has high ability in handling reservoir operation is crucial. The present study developed artificial intelligence model, called Shark Machine Learning Algorithm (SMLA) to provide optimal operational rules. The major objective for the proposed model is minimizing the deficit volume between water releases and the irrigation water demand. The current study compared the performance of the SML model with popular evolutionary computing methods, namely Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). The proposed models have been utilized of finding the optimal policies to operate Timah Tasoh Dam, which is located in Malaysia. The study utilized considerable statistical indicators to explore the efficiency of the models. The simulation period showed that SMLA approach outperforms both of conventional algorithms. The SMLA attained high Reliability and Resilience (Rel. = 0.98%, Res. = 50%) and minimum Vulnerability (Vul. = 21.9 of demand). It is demonstrated that shark machine learning algorithm would be a promising tool in handling the long-term optimization problem in operation a reservoir system.     

Chaotic Evolutionary Algorithms for Multi-Reservoir Optimization

The water sharing dispute in a multi-reservoir river basin forces the water resources planners to have an integrated operation of multi-reservoir system rather than considering them as a single reservoir system. Thus, optimizing the operations of a multi-reservoir system for an integrated operation is gaining importance, especially in India. Recently, evolutionary algorithms have been successfully applied for optimizing the multi-reservoir system operations. The evolutionary optimization algorithms start its search from a randomly generated initial population to attain the global optimal solution. However, simple evolutionary algorithms are slower in convergence and also results in sub-optimal solutions for complex problems with hardbound variables. Hence, in the present study, chaotic technique is introduced to generate the initial population and also in other search steps to enhance the performance of the evolutionary algorithms and applied for the optimization of a multi-reservoir system. The results are compared with that of a simple GA and DE algorithm. From the study, it is found that the chaotic algorithm with the general optimizer has produced the global optimal solution (optimal hydropower production in the present case) within lesser generations. This shows that coupling the chaotic algorithm with evolutionary algorithm will enrich the search technique by having better initial population and also converges quickly. Further, the performances of the developed policies are evaluated for longer run using a simulation model to assess the irrigation deficits. The simulation results show that the model satisfactorily meets the irrigation demand in most of the time periods and the deficit is very less.     

Single Reservoir Operating Policies Using Genetic Algorithm

To obtain optimal operating rules for storage reservoirs, large numbers of simulation and optimization models have been developed over the past several decades, which vary significantly in their mechanisms and applications. As every model has its own limitations, the selection of appropriate model for derivation of reservoir operating rule curves is difficult and most often there is a scope for further improvement as the model selection depends on data available. Hence, evaluation and modifications related to the reservoir operation remain classical. In the present study a Genetic Algorithm model has been developed and applied to Pechiparai reservoir in Tamil Nadu, India to derive the optimal operational strategies. The objective function is set to minimize the annual sum of squared deviation form desired irrigation release and desired storage volume. The decision variables are release for irrigation and other demands (industrial and municipal demands), from the reservoir. Since the rule curves are derived through random search it is found that the releases are same as that of demand requirements. Hence based on the present case study it is concluded that GA model could perform better if applied in real world operation of the reservoir.     

Optimal Reservoir Release Policy Considering Heterogeneity of Command Area by Elitist Genetic Algorithm

A number of models with conventional optimization techniques have been developed for optimization of reservoir water release policies. However these models are not able to consider the heterogeneity in the command area of the reservoir appropriately, due to non linear nature of the processes involved. The optimization model based on genetic algorithm (GA) can deal with the non linearity due to its inherent ability to consider complex simulation model as evaluation function for optimization. GA based models available in literature generally minimize the water deficits and do not optimize the total net benefits through optimal reservoir release policies. The present study focuses on optimum releases from the reservoir considering heterogeneity of the command area and responses of the command area to the releases instead of minimizing only the reservoir storage volumes. An optimization model has been developed for the reservoir releases based on elitist GA approach considering the heterogeneity of the command area. The developed model was applied to Waghad irrigation project in upper Godavari basin of Maharashtra, India. The results showed that 19% increase in the total net benefits could be possible by adopting the proposed water release policy over the present practice keeping same distribution of area under different crops. The model presented in this study can also optimize the crop area under irrigation. It is found that irrigated area can be increased to 50% of ICA (Irrigable Command Area) from the existing 23% with resulting addition to total net benefits by 31%. The effect of adopting the proposed irrigation schedule and increased irrigation areas would be to increase the net benefits to existing farmers.     

Ant Colony Optimization for Multi-Purpose Reservoir Operation

In this paper a metaheuristic technique called Ant Colony Optimization (ACO) is proposed to derive operating policies for a multi-purpose reservoir system. Most of the real world problems often involve non-linear optimization in their solution with high dimensionality and large number of equality and inequality constraints. Often the conventional techniques fail to yield global optimal solutions. The recently proposed evolutionary algorithms are also facing problems, while solving large-scale problems. In this study, it is intended to test the usefulness of ACO in solving such type of problems. To formulate the ACO model for reservoir operation, the problem is approached by considering a finite time series of inflows, classifying the reservoir volume into several class intervals, and determining the reservoir release for each period with respect to a predefined optimality criterion. The ACO technique is applied to a case study of Hirakud reservoir, which is a multi-purpose reservoir system located in India. The multiple objectives comprise of minimizing flood risks, minimizing irrigation deficits and maximizing hydropower production in that order of priority. The developed model is applied for monthly operation, and consists of two models viz., for short-time horizon operation and for long-time horizon operation. To evaluate the performance of ACO, the developed models are also solved using real coded Genetic Algorithm (GA). The results of the two models indicate that ACO model performs better, in terms of higher annual power production, while satisfying irrigation demands and flood control restrictions, compared to those obtained by GA. Finally it is found that ACO model outperforms GA model, especially in the case of long-time horizon reservoir operation.     

Optimizing Reservoir Operation Policy Using Chance Constraint Nonlinear Programming for Koga Irrigation Dam,Ethiopia

One of typical problems in water resources system modeling is derivation of optimal operating policy for reservoir to ensure water is used more efficiently. This paper introduces optimization analysis to determine monthly reservoir operating policies for five scenarios of predetermined cropping patterns for Koga irrigation scheme, Ethiopia. The objective function of the model was set to minimize the sum of squared deviation (SSD) from the desired targeted supply. Reservoir operation under different water availability and thresholds of irrigation demands has been analyzed by running a chance constraint nonlinear programming model based on uncertain inflow data. The model was optimized using Microsoft Excel Solver. The lowest SSD and vulnerability, and the highest volumetric reliability were gained at irrigation deficit thresholds of 20 % under scenario I, 30 % under scenario II, III and V, and at 40 % under scenario IV when compensation release is permitted for downstream environment. These thresholds of deficits could be reduced by 10 % for all scenarios if compensation release is not permitted. In conclusion the reservoir water is not sufficient enough to meet 100 % irrigation demand for design command areas of 7,000 ha. The developed model could be used for real time reservoir operation decision making for similar reservoir irrigation systems. In this specific case study system, attempt should be made to evaluate the technical performance of the scheme and introduce a regulated deficit irrigation application.     

Investment in Irrigation Systems under Precipitation Uncertainty

Efficient agricultural water management is indispensable in meeting future food demands. The European Water Framework Directive promotes several measures such as the adoption of adequate water pricing mechanisms or the promotion of water-saving irrigation technologies. We apply a stochastic dynamic programming model (SDPM) to analyze a farmer??s optimal investment strategy to adopt a water-efficient drip irrigation system or a sprinkler irrigation system under uncertainty about future production conditions, i.e. about future precipitation patterns. We assess the optimal timing to invest into either irrigation system in the planning period 2010 to 2040. We then investigate how alternative policies, (a) irrigation water pricing, and (b) equipment subsidies for drip irrigation, affect the investment strategy. We perform the analysis for the semi-arid agricultural production region Marchfeld in Austria, and use data from the bio-physical process simulation model EPIC (Environmental Policy Integrated Climate) which takes into account site and management related characteristics as well as weather parameters from a statistical climate change model. We find that investment in drip irrigation is unlikely unless subsidies for equipment cost are granted. Also water prices do not increase the probability to adopt a drip irrigation system, but rather delay the timing to invest into either irrigation system.     

Chance Constrained Optimal Design of Composite Channels Using Meta-Heuristic Techniques

Optimal design of irrigation channels has an important role in planning and management of irrigation projects. The input parameters used in design of irrigation channels are prone to uncertainty and may result in failure of channels. To improve the overall reliability and cost effectiveness, optimal design of composite channels is performed as a chance constrained problem in this study. The models are developed to minimize the total cost, while satisfying the specified probability of the channel capacity being greater than the design flow. The formulated model leads to a highly non-linear and non-convex optimization problem having multimodal behavior. In this paper, the usefulness of two meta-heuristic search algorithms such as Genetic Algorithms (GA) and Particle Swarm Optimization (PSO) are investigated to obtain the optimal solutions. Two site specific cases of restricted top width and restricted flow depth are also analyzed. It is found that both the algorithms performing quite well in giving optimal solutions and handling the additional constraints.     

Integrated Artificial Neural Network (ANN) and Stochastic Dynamic Programming (SDP) Model for Optimal Release Policy

Complexicity in reservoir operation poses serious challenges to water resources planners and managers. These challenges of water reservoir operation are illustrated using a simulation to aid the development of an optimal operation policy for dam and reservoir. To achieve this, a Comprehensive Stochastic Dynamic Programming with Artificial Neural Network (SDP-ANN) model were developed and tested at Sg. Langat Reservoir in Malaysia. The nonlinearity of the natural physical processes was a major problem in determining the simulation of the reservoir parameters (elevation, surface-area, storage). To overcome water shortages resulting from uncertainty, the SDP-ANN model was used to evaluate the input variable and the performance outcome of the Model were compared with the Stochastic Dynamic Programming integrated with auto-regression (SDP-AR) model. The objective function of the models was set to minimize the sum of squared deviation from the desired targeted supply. Comparison result on the performance between SDP-AR model policy with SDP-ANN model found that the SDP-ANN model is a reliable and resilience model with a lesser supply deficit. The study concludes that the SDP-ANN model performs better than the SDP-AR model in deriving an optimal operating policy for the reservoir.     

Reservoir Operation by a New Evolutionary Algorithm: Kidney Algorithm

This article shows an application of a new algorithm, called kidney algorithm, for reservoir operation which employs three different operators, namely filtration, secretion, and excretion that lead to faster convergence and more accurate solutions. The kidney algorithm (KA) was used for generating the optimal operation of a reservoir namely; Aydoghmoush dam in eastern Azerbaijan province in Iran whose purpose was to decrease irrigation deficit downstream of the dam. Results from the algorithm were compared with those by other evolutionary algorithms, including bat (BA), genetic (GA), particle swarm (PSO), shark (SA), and weed algorithms (WA). The results showed that the kidney algorithm provided the best performance against the other evolutionary algorithms. For example, the computational time for the KA was 3 s, 2 s, 4 s, 6 s and 3 s less than BA, SA, GA PSA and WA, respectively. Also, the objective function for the optimization problem was the minimization of the irrigation deficits and its value for the KA was 55%, 28%, 52%, 44 and 54% less than GA, SA, WA, BA and PSA, respectively. Also, the different performance indexes showed the superiority of the KA compared to the other algorithms. For example, the root mean square error for the KA was 74%, 61%, 68%, 33 and 54% less than GA, SA, WA, BA and PSA, respectively. Different multi criteria decision models were used to select the best models. The results showed that the KA achieved the first rank for the optimization problem and thus, it shows a high potential to be applied for different problems in the field of water resources management.     

Sustainable Irrigation Planning with Imprecise Parameters under Fuzzy Environment

The present paper proposes a model of Multi Objective Fuzzy Linear Programming (MOFLP) based on Fuzzy Parametric Programming (FPP) to solve the problem of optimal cropping pattern in an irrigation system. It has been found that in order to solve the problem of uncertainty in the planning of sustainable irrigation, the concept of fuzzy logic has been in practice for long and was being systematically applied either to the case of fuzzy objectives alone or to case of fuzzy objectives with fuzzy resources. There has not been reported a single case of either formulation and application of MOFLP model for the planning of irrigation making use of fuzzy objective function coefficients, fuzzy technological coefficients and fuzzy resources. The approach presented in the MOFLP model attempts to consider the fuzziness of all the coefficients of a mathematical model, as they present themselves in the real life situations. The present model takes into account the experience, information and expectations of the Decision Maker (DM). The objective of the model is to maximize simultaneously four objective functions viz. the Net Benefits (NB), Crop Production (CP), Employment Generation (EG) and Manure Utilization (MU). The model proposed takes into consideration the fuzziness involved in the coefficient of objective functions, technological coefficients and stipulations. The model intends to develop a program of sustainable irrigation planning for the Jayakwadi Project, Stage-I, located in the State of Maharashtra, India. The optimal cropping pattern has been obtained for five different strategies. The results finally obtained through the fifth strategy appear realistic, promising and effective as they involve the consideration of the uncertainty contained in coefficient of objective functions, technological coefficients and stipulations simultaneously. The model may be applied to any irrigation project with a view to utilize the resources available optimally and deal with the problem of uncertainty in realistic ways in solving real life problems.     

Reliability-based Operation of Reservoirs Using Combined Monte Carlo Simulation Model and a Novel Nature-inspired Algorithm

One of the critical issues in surface water resources management is the optimal operation of dam reservoirs. In recent decades, meta-heuristics algorithms have gained attention as a powerful tool for finding the optimal program for the dam reservoir operation. Increasing demand due to population growth and lack of precipitation for reasons such as climate change has caused uncertainties in the affecting parameters on the planning of reservoirs, which invalidates the operational plans of these reservoirs. In this study, a novel optimization algorithm with the combination of genetic algorithm (GA) and multi-verse optimizer (MVO) called multi-verse genetic algorithm (MVGA) has been developed to solve the optimal dam reservoir operation issue under influence of the joint uncertainties of inflow, evaporation and demand. After validating the performance of MVGA by solving several benchmark functions, MVGA was used to find the optimal operation program of the Amirkabir Dam reservoir in 132 months, in both deterministic and probabilistic states. Minimizing the deficit between downstream demand and release from the reservoir during the operation period was considered as the objective function. Also, the limitations of the reservoir continuity equation, storage volume, and reservoir release equation were applied to the objective function. For modeling the effect of uncertainty, Monte Carlo simulation (MCS) is coupled to MVGA. The results of model implementations showed that the MVGA-MCS model with the best value of the objective function equal to 26 in the 1st rank and MVGA, MVO, and GA, with 15%, 34%, and 46% increase in the value of the objective function compared to the MVGA-MCS stood in the second to fourth ranks, respectively. Also, the results of the resiliency, and vulnerability indices of the reservoir operation showed that MVGA-MCS and MVGA models have better performance than other models.

     

Evaluation of Crop Models for Simulating and Optimizing Deficit Irrigation Systems in Arid and Semi-arid Countries Under Climate Variability

The variability of fresh water availability in arid and semi-arid countries poses a serious challenge to farmers to cope with when depending on irrigation for crop growing. This has shifted the focus onto improving irrigation management and water productivity (WP) through controlled deficit irrigation (DI). DI can be conceived as a strategy to deal with these challenges but more knowledge on risks and chances of this strategy is urgently needed. The availability of simulation models that can reliably predict crop yield under the influence of soil, atmosphere, irrigation, and agricultural management practices is a prerequisite for deriving reliable and effective deficit irrigation strategies. In this context, this article discusses the performance of the crop models CropWat, PILOTE, Daisy, and APSIM when being part of a stochastic simulation-based approach to improve WP by focusing primarily on the impact of climate variability. The stochastic framework consists of: (i) a weather generator for simulating regional impacts of climate variability; (ii) a tailor-made evolutionary optimization algorithm for optimal irrigation scheduling with limited water supply; and (iii) the above mentioned models for simulating water transport and crop growth in a sound manner. The results present stochastic crop water production functions (SCWPFs) that can be used as basic tools for assessing the impact on the risk for the potential yield due to water stress and climate variability. Example simulations from India, Malawi, France and Oman are presented and the suitability of these crop models to be employed in a framework for optimizing WP is evaluated.     

Reliability-Based Pipe Size Computation of On-Demand Irrigation Systems

The variability of flow regimes in on demand pressurized irrigation systems leads to uncertainty in head at the nodes affecting the system performance, and thus it should be considered when designing and/or rehabilitating water distribution systems. Based on these considerations a new approach for the optimization of on demand pressurized irrigation systems is presented combining the minimization of cost with the maximization of reliability taking into account the stochastic variability of the flows into each section of the network. The new model, Clément and the cumulated random generated discharges model (FAO model) were applied to three pressurized irrigation networks of different dimensions (large, medium and small) operating on demand in Southern Italy. The optimization algorithm used in all the cases is the Labye iterative discontinuous method, a formulation of the dynamic programming. The results of the different models were compared showing that the cost of the optimal network calculated using the new model was reduced by more than 20%, without any significant decrease of the system reliability or reduction of the network capacity.     

A Fuzzy Stochastic Dynamic Nash Game Analysis of Policies for Managing Water Allocation in a Reservoir System

In this paper a fuzzy dynamic Nash game model of interactions between water users in a reservoir system is presented. The model represents a fuzzy stochastic non-cooperative game in which water users are grouped into four players, where each player in game chooses its individual policies to maximize expected utility. The model is used to present empirical results about a real case water allocation from a reservoir, considering player (water user) non-cooperative behavior and also same level of information availability for individual players. According to the results an optimal allocation policy for each water user can be developed in addition to the optimal policy of the reservoir system. Also the proposed model is compared with two alternative dynamic models of reservoir optimization, namely Stochastic Dynamic Programming (SDP) and Fuzzy-State Stochastic Dynamic programming (FSDP). The proposed modeling procedures can be applied as an appropriate tool for reservoir operation, considering the interaction among the water users as well as the water users and reservoir operator.     

Irrigation Planning Under Uncertainty—A Multi Objective Fuzzy Linear Programming Approach

The problem of irrigation planning becomes more complex by considering an uncertainty. The uncertainties can be tackled by formulating the problem of irrigation planning as Fuzzy Linear Programming (FLP). FLP models can incorporate the scenario of real world problem. In the present study, Multi Objective Fuzzy Linear Programming (MOFLP) irrigation planning model is formulated for deriving the optimal cropping pattern plan for the case study of Jayakwadi project in the Godavari river sub basin in Maharashtra State, India. Four conflicting objectives are considered such as Net Benefits (NB), Crop/Yield Production (CP), Employment Generation/Labour Requirement (EG) and Manure Utilization (MU). Four different cases are considered to incorporate the uncertainty in MOFLP model. To include the uncertainty in irrigation planning problem only objectives are taken as fuzzy and constraints are crisp in nature in Case-I. To consider the uncertainty involved in availability of resources, in Case-II the stipulations are fuzzy. The technological coefficients are fuzzy in Case-III. The Case-IV includes both technological coefficients and stipulations fuzzy. The level of satisfaction (λ) works out to be 0.58, 0.50, 0.50 and 0.28 respectively for Case-I to IV. The results obtained in Case-IV are more realistic and promising as it involves the uncertainty in technological coefficients and stipulations simultaneously.     

Application of Genetic Algorithms and Artificial Neural Networks in Conjunctive Use of Surface and Groundwater Resources

Abstract

In this paper, a methodology for conjunctive use of surface and groundwater resources is developed using the combination of the Genetic Algorithms (GAs) and the Artificial Neural Networks (ANN). Water supply to agricultural demands, reduction of pumping costs and control of groundwater table fluctuations are considered in the objective function of the model. In the proposed model, the results of MODFLOW groundwater simulation model are used to train an ANN. The ANN as groundwater response functions is then linked to the GA based optimization model to develop the monthly conjunctive use operating policies. The model is applied to the surface and groundwater allocation for irrigation purposes in the southern part of Tehran. A new ANN is also trained and checked for developing the real-time conjunctive use operating rules.

The results show the significance of an integrated approach to surface and groundwater allocation in the study area. A simulation of the optimal policies shows that the cumulative groundwater table variation can be reduced to less than 4 meters from the current devastating condition. The results also show that the proposed model can effectively reduce the run time of the conjunctive use models through the composition of a GA-based optimization and a ANN-based simulation model.