Optimal Reservoir Operation Research Articles

Turkey’s Hydropower Potential in the Near Future and the Possible Impacts of Climate Change—A Case Study of the Euphrates–Tigris Basin

Hydropower is becoming an important renewable energy source in Turkey, but the ever-changing atmospheric and climatic conditions of Turkey make it very difficult to be projected efficiently. Thus, an efficient estimation technique is crucial for it to be adopted as a reliable energy source in the future. This study evaluates Turkey’s hydropower potential in the Euphrates–Tigris Basin under changing climatic conditions. We adapted an empirical equation to model reservoir outflows, considering the site-specific characteristics of 14 major dams. Initial results from employing a model with a constant empirical coefficient, α, yielded moderate predictive accuracy, with R2 values ranging from 0.289 to 0.612. A polynomial regression identified optimal α values tailored to each dam’s surface area, significantly improving model performance. The adjusted α reduced predictive bias and increased R2 values, enhancing forecast reliability. Seasonal analysis revealed distinct hydropower trends: Ataturk Dam showed a notable decrease of 5.5% in hydropower generation up to 2050, while Birecik and Keban Dams exhibited increases of 2.5% and 2.2%, respectively. By putting these discoveries into practice, water resource management may become more robust and sustainable, which is essential for meeting Turkey’s rising energy needs and preparing for future climatic challenges. This study contributes valuable insights for optimizing reservoir operations, ensuring long-term hydropower sustainability, and enhancing the resilience of water resource management systems globally.

Comparative study of reservoir operations using TLBO, PSO and DE optimization techniques: an experiment on the Hirakud reservoir, Odisha, India

ABSTRACT With the enduring progression of high population density, social economy and requirements of water supply, irrigation and hydropower, the water resource scarcity problem has been exacerbated in Odisha. Hence, the judicial operation of the Hirakud reservoir, which is considered the lifeline of Odisha, has become appreciably essential. Among various optimization techniques, metaheuristic algorithms are advanced techniques which can be applied for the optimal operation of a water reservoir. In this work, three metaheuristic algorithm-based optimization techniques, the Particle Swarm Optimization (PSO), Differential Evolution (DE) and Teaching–Learning-Based Optimization (TLBO) algorithms, are applied for optimal water management of the Hirakud reservoir. From the result, it was found that the efficiency of TLBO for irrigation release during the non-monsoon period was 99.45% compared with PSO with an efficiency of 97.3% and DE with an efficiency of 95.6%. The efficiency of TLBO for hydropower generation was 99.6%, whereas for PSO it was 98.8% and for DE it was 98.5%. It is found from the above experiment that TLBO showed a better performance than PSO and DE optimization techniques for the water management of the Hirakud reservoir. These metaheuristic techniques will provide suitable guidance for reservoir operations at times when the presence of experts is a must but they may not be available.

Optimization of reservoir operation by sine cosine algorithm: A case of study in Algeria

The optimal operation of the reservoir has vital importance in water engineering. In the presented article, a new optimization method, named sine cosine algorithm (SCA) was employed to obtain operating policy for an irrigation system. The SCA was utilized for the monthly operation of the Boukerdane Dam placed in the north of Algeria. The fitness function was the minimization of the total shortage for the studied period. Three scenarios considering three different seasons of inflow (dry, normal and wet) are used to optimize the reservoir system’s operation. The SCA outputs were compared with particle swarm optimization (PSO) and kidney algorithm (KA). The outcomes indicated that the SCA surpassed the PSO and KA in convergence rate. The general results indicated the low speed of KA and PSO in achieving convergence. The results indicated that the highest RES (resiliency index), SUS (sustainability index) and REL (reliability index) achieved by the SCA were 65, 86 and 92 %, respectively. Comparing the third scenario with the first and second scenarios, it was observed that the third scenario (wet seasons) improved the results.

MILP and PSO approaches for solving a hydropower reservoirs intraday economic optimization problem

AbstractShort-term hydropower generation with several water reservoirs requires deciding, for each moment in time, the volume of water (outflow) that is released from every reservoir to be turbined and generate energy. Knowing the price of energy at every time period, the objective is to maximize the income earned from the generated energy. In this paper, we present (1) a Hydropower Reservoirs Operation Optimization problem with a higher level of detail than those found in the literature, encompassing temporal delays, water hammer effects, and increased temporal discretization, among others features, and (2) two distinct approaches for addressing this problem: MILP and PSO. These methods are compared across instances of varying nature to evaluate their performance. We make our code available on GitHub: https://github.com/baobabsoluciones/flowing-basin.

Rapid identification of back-flow in three distributaries of Jingjiang Reach, middle Yangtze River, China

Unusual back-flow events occurred in 2015 and 2017 at the near-mainstream ends of the three distributaries of the Jingjiang Reach, middle Yangtze River, China. It is presently challenging to precisely quantify back-flow discharges at river distributary ends where normal-flow discharges are measured chronically. This problem is more complicated at the ends of the three Jingjiang distributaries, each of which is more than 100 km long, and thus favors the generation of different back-flow forms. We use empirical formulae based on the Saint-Venant Equation and the Fall Index Method to quantify impacts on flow discharges at the two ends of each distributary of three key hydrodynamic variables: (1) water discharge at Zhicheng hydrological station, (2) total discharge from the four southwest tributary rivers, and (3) water levels at the two ends of the distributaries. We use the variable control approach to identify the contribution ratios of natural and human driving factors to flow discharges at the two ends of the distributaries during five historical stages. The empirical formulae provide reasonable results, with high values of correlation coefficients (R2 = 0.84 ∼ 0.98) obtained, demonstrating that the total water discharge from the four southwest tributary rivers is negligible. These formulae perform well in reproducing observed flow discharges at the two ends of the distributaries, especially the maximum discharges and cumulative volumes for normal-flow and back-flow events. Three charts are devised for different flow conditions at the two ends, from which back-flow events can be rapidly identified. Before impoundment of the Three Gorges Dam (TGD), riverbed adjustment caused by the lower Jingjiang Cutoff Projects and the Gezhou Dam contributed the most to flow discharge changes (≥50 %) at the near-mainstream ends of the distributaries. Afterwards, precipitation dominated (>60 %). In the future, back-flow events are expected only to occur at Mituosi (i.e., Taipingkou) and Nanxian hydrological stations, for which optimal reservoir operation and water resource utilization could remedy the back-flow crisis. Use of climate models and more advanced technologies such as remote sensing and GIS are recommended to enhance the reliability of predictions.

Improved Multi-objective Butterfly Optimization Algorithm and its Application in Cascade Reservoirs Optimal Operation Considering Ecological Flow

Improved Multi-objective Butterfly Optimization Algorithm and its Application in Cascade Reservoirs Optimal Operation Considering Ecological Flow

Constraint programming for reservoir operation optimization of Bhumibol dam

The modern constraint programming (CP) was adopted to minimize water scarcity and excessive water which are the critical issues in reservoir operation of Bhumibol Dam (BB) to solve consecutive droughts in the Chao Phraya River Basin (CPYRB), Thailand. The situations of the severe droughts have been frequently occurred in a broad area of CPYRB due to an extremely low rainfall leading to unusually low water levels and inflow in major reservoirs. Therefore, multi-objective optimization was conducted to characterize the actual operation and physical reservoir system of BB Dam. Two different CP models with seasonal and yearly constraints were manipulated using MiniZinc programming language and the constraint solver IPOPT to find the optimal daily release scheme from 2000 to 2018 of BB Dam. The potential of downstream flow conditions was also considered to partially supply downstream water demand and store savable water in a reservoir for subsequent use during possible future droughts. The results reveal that CP models can diminish some extent of yearly reservoir release, while daily long-term release scheme conforms well with the actual operation particularly during dry and wet seasons in specific drought years. These mean that amount of reservoir water of approximately 47.12–103.83 MCM/year can be saved and stored in reservoir for subsequent use in CPYRB when CP models are deployed. This results in a reservoir storage increase of 7.10–7.94% at the end of the wet season for seasonal and yearly CP models, respectively. When potential side flow is considered, the increase climbs up to 10.49%. This envisages the higher possibility of supplying reservoir water for agricultural water needs over the dry season in the Greater Chao Phraya Irrigation Scheme. As the potential hydraulic head is subject to increased reservoir water storage, therefore, the increase in hydropower production is definitely found of ranging from + 6.10% to + 13.79% by these two sorts of CP models. In addition, handling with huge volumes of flood and drought by CP models is well managed as the specific release constraints during refilled and drawdown periods are used to define the constraint satisfaction problem.

Field monitoring and modelling of sediment transport, hydraulics and hydroabrasion at Sediment Bypass Tunnels

Sediment Bypass Tunnels (SBTs) are proven to be an effective measure to reduce or even stop reservoir sedimentation by bypassing sediment laden flows around reservoir dams to the downstream river reach. They are mostly used in Switzerland, Japan, and Taiwan. However, hydraulic and sedimentological operating conditions and the resistance of the invert materials against hydroabrasive erosion affect their cost-effectiveness. Hydroabrasion is a pressing issue at SBTs, other hydraulic structures and steep bedrock rivers exposed to high sediment transport rates under supercritical flow conditions. The present study was therefore conducted to address this issue by aiming at improving knowledge on abrasion mechanics and calibrating a mechanistic saltation abrasion model enhanced by Demiral-Yüzügüllü (2021). To this end, the abrasion resistance of fourteen different invert materials installed at Solis, Pfaffensprung and Runcahez SBTs in Switzerland was quantified by annual 3D laser scanning and the hydraulic conditions and sediment transport rates were regularly monitored between 2017 and 2021. The analysis of invert scans and hydraulic conditions revealed that Prandtl’s first and second kinds of secondary currents occurring in the bends and straight sections of the SBTs, respectively, and the observed abrasion patterns were strongly interrelated. The tested potassium aluminate cement and steel fibre concretes, granite, cast basalt and steel plates had better abrasion resistance against impact of sediment-laden flows compared to other materials. Sediment mineralogical composition i.e., bulk hardness relative to the invert material properties significantly affected hydroabrasion. The enhanced abrasion prediction model was calibrated with the present data and a quasi-constant abrasion coefficient of kv = (4.8 ± 2.2) × 104 was obtained. The enhanced model is well-suited for both laboratory and field scales. The present findings will contribute to the sustainable utilization and operational safety of hydraulic structures, optimization of SBT and reservoir operations regarding bypassing efficiency and reservoir lifetime and modelling of bedrock river erosion.

Optimal operation of the dam reservoir in real time based on generalized structure of group method of data handling and optimization technique

The historical data on water intake into the reservoir is collected and used within the framework of a deterministic optimization method to determine the best operating parameters for the dam. The principles that have been used to extract the best values of the flow release from the dam may no longer be accurate in the coming years when the inflow to dams will be changing, and the results will differ greatly from what was predicted. This represents this method’s main drawback. The objective of this study is to provide a framework that can be used to guarantee that the dam is running as efficiently as possible in real time. Because of the way this structure is created, if the dam’s inflows change in the future, the optimization process does not need to be repeated. In this case, deep learning techniques may be used to restore the ideal values of the dam’s outflow in the shortest amount of time. This is achieved by accounting for the environment’s changing conditions. The water evaluation and planning system simulator model and the MOPSO multi-objective algorithm are combined in this study to derive the reservoir’s optimal flow release parameters. The most effective flow discharge will be made feasible as a result. The generalized structure of the group method of data handling (GSGMDH), which is predicated on the results of the MOPSO algorithm, is then used to build a new model. This model determines the downstream needs and ideal release values from the reservoir in real time by accounting for specific reservoir water budget factors, such as inflows and storage changes in the reservoir. Next, a comparison is drawn between this model’s performance and other machine learning techniques, such as ORELM and SAELM, among others. The results indicate that, when compared to the ORELM and SAELM models, the GSGMDH model performs best in the test stage when the RMSE, NRMSE, NASH, and R evaluation indices are taken into account. These indices have values of 1.08, 0.088, 0.969, and 0.972, in that order. It is therefore offered as the best model for figuring out the largest dam rule curve pattern in real time. The structure developed in this study can quickly provide the best operating rules in accordance with the new inflows to the dam by using the GSGMDH model. This is done in a way that makes it possible to manage the system optimally in real time.

Multi-Objective Optimal Operation Decision for Parallel Reservoirs Based on NSGA-II-TOPSIS-GCA Algorithm: A Case Study in the Upper Reach of Hanjiang River

The parallel reservoirs in the upper reach of the Hanjiang River are key projects for watershed management, development, and protection. The optimal operation of parallel reservoirs is a multiple-stage, multiple-objective, and multiple-decision attributes complex decision problem. Taking Jiaoyan–Shimen parallel reservoirs as an example, a method of multi-objective optimal operation decision of parallel reservoirs (MOODPR) was proposed. The multi-objective optimal operation model (MOOM) was constructed. The new algorithm coupling NSGA-II, TOPSIS, and GCA was used to solve the MOODPR problem. The method of MOODPR was formed by coupling problem identification, model construction, an optimization solution, and scheme evaluation. The results show that (1) combining the Euclidean distance with the grey correlation degree to construct a new hybrid closeness degree makes the multi-attribute decision making method more scientific and feasible. (2) The NSGA-II-TOPSIS-GCA algorithm is applied to obtain decision schemes, which provide decision support for management. (3) It can be seen from the Pareto chart that for the Jiaoyan–Shimen parallel reservoirs, the comprehensive water supply was negatively related to ecology. (4) The comprehensive water supply and ecological AAPFD value in the extraordinarily dry year was 4.212 × 108 m3 and 4.953. The number of maximum continuous water shortage periods was 4 and 6. The maximum ten-day water shortage was 4.46 × 107 m3 and 2.3 × 106 m3. The research results provide technical support and reference value to multi-objective optimal operation decisions for parallel reservoirs in the upper reach of the Hanjiang River.

Optimizing reservoir operation incorporating ecological demand and stability requirement

AbstractThe construction and operation of large‐scale reservoirs alter the natural river flow regimes and destroy the original suitable reproduction and living conditions of aquatic organisms. Especially during the spawning season, the fluctuating water flow process is particularly important for fish reproduction. A bi‐objective optimization model considering both the environmental flow demand for fish spawning and the stability of hydropower output is established in this study and solved by one of the most widely used algorithms for multi‐objective problems, the Non‐Dominated Sorted Genetic Algorithm‐II. The derived Pareto Front can be used to guide reservoir operation during fish spawning seasons. The model proposed in this paper is applied to the Yangqu Reservoir, currently under construction on the upper Yellow River in China. The results show that the stability of hydropower output can still be improved by 4.80% to 10.56%, even if the environmental flow demand is taken into account. Even for the scheme preferring ecological recovery, the stability of hydropower output can be increased by more than 0.85%.

Synoptic Analysis of Flood-Causing Rainfall and Flood Characteristics in the Source Area of the Yellow River

The source area of the Yellow River (SAYR) in China is an important water yield and water-conservation area in the Yellow River. Understanding the variability in rainfall and flood over the SAYR region and the related mechanism of flood-causing rainfall is of great importance for the utilization of flood water resources through the optimal operation of cascade reservoirs over the upper Yellow River such as Longyangxia and Liujiaxia, and even for the prevention of flood and drought disasters for the entire Yellow River. Based on the flow data of Tangnaihai hydrological station, the rainfall data of the SAYR region and NCEP-NCAR reanalysis data from 1961 to 2020, three meteorological conceptual models of flood-causing rainfall—namely westerly trough type, low vortex shear type, and subtropical high southwest flow type—are established by using the weather-type method. The mechanism of flood-causing rainfall and the corresponding flood characteristics of each weather type were investigated. The results show that during the process of flood-causing rainfall, in the westerly trough type, the mid- and high-latitude circulation is flat and fluctuating. In the low vortex shear type, the high pressures over the Ural Mountains and the Okhotsk Sea are stronger compared to other types in the same period, and a low vortex shear line is formed in the west of the SAYR region at the low level. The rain is formed during the eastward movement of the shear line. In the subtropical high southwest flow type, the low trough of Lake Balkhash and the subtropical high are stronger compared to other types in the same period. Flood-causing rainfall generally occurs in areas with low-level convergence, high-level negative vorticity, low-level positive vorticity, convergence of water vapor flux, a certain amount of atmospheric precipitable water, and low-level cold advection. In terms of flood peak increment and the maximum accumulated flood volume, the westerly trough type has a long duration and small flood volume, and the low vortex shear type and the subtropical high southwest flow type have a short duration and large flood volume.

Study on the constraint handling method for high-dimensional optimization of cascade reservoirs

Multicomplex constraints often need to be considered in the optimal operation of cascade reservoirs with high-dimensional decision variables, making it difficult for traditional optimization methods and modern intelligent algorithms to solve such problems. Therefore, this study proposes a constraint handling method combining a penalty function nested DPSA-POA and an intelligent algorithm, to determine the optimal flood control operation of cascade reservoirs in the middle reaches of the Ganjiang River, a problem with decision variables of up to 2196 dimensions. The results indicate that the constraint handling method proposed in this paper can solve high-dimensional optimization problems in three modes: continuous nesting (Mode 1), optimization with intelligence after obtaining a feasible solution (Mode 2), and optimization with the DPSA-POA after obtaining a feasible solution (Mode 3). Of the three modes, Mode 2 has the highest accuracy, but its calculation time is approximately 10 h. Although the accuracy of Mode 3 is slightly worse, can only achieve 98%–99% of Mode 2, its calculation time is only approximately 1–3 h. The comprehensive performance of Mode 1 is poor, the convergence accuracy can only reach 97% of Mode 2, which corresponds to a calculation time of approximately 4–6 h. The existing superiority of feasibility (SF), stochastic ranking (SR), penalty function (PF), ε-constraint (EC) and adaptive ε-constraint (Adaptive EC) methods cannot converge to a feasible solution stably, and the accuracy of the results of these methods under the condition of obtaining a feasible solution is significantly lower than that of Mode 2 and Mode 3, reaching only 95%.

An Alternative Approach Using the Firefly Algorithm and a Hybrid Method Based on the Artificial Bee Colony and Cultural Algorithm for Reservoir Operation

In reservoir operation rule curves, it is necessary to apply rule curves to guide long-term reservoir management. This study proposes an approach to optimizing reservoir operation rule curves (RORCs) using intelligent optimization techniques from the firefly algorithm (FA) and a unique combination method utilizing the artificial bee colony and cultural algorithm (ABC-CA). The aim is to establish a connection with the simulation model to determine the optimal RORCs for flood control. The proposed model was used to determine the optimal flood control RORC for the Nam-Oon Reservoir (NOR) in northeastern Thailand. A minimum frequency and minimum average of excess water were provided as an objective function for assessing the efficiency of the search process. The evaluation of the effectiveness of flood control RORCs involved expressing water scarcity and excess water situations in terms of frequency, magnitude, and duration using historical inflow data synthesized from 1000 events. The results demonstrated that when using the obtained RORC to simulate the NOR system for reducing flooding in long-term operations, excess water scenarios were smaller than those using the current RORC. The results showed that the excess water scenario using the RORC obtained from the proposed model can reduce the excess water better than the current RORC usage scenario. In decreasing flood situations, the newly acquired RORC from the suggested FA and ABC-CA models performed better than the current RORC.

A Simulation–Optimization System to Assess Dam Construction with a Focus on Environmental Degradation at Downstream

AbstractThe present study proposes and evaluates an integrated framework to assess dam construction and removal, encompassing the simulation of downstream river habitats and reservoir operation in three distinct statuses: conventional reservoir operation optimization, optimal release considering environmental aspects within the optimization model, and natural flow conditions. Fuzzy physical habitat simulation was employed to assess physical habitats, while an ANFIS-based model was utilized to simulate thermal tension and dissolved oxygen tension at downstream habitats. Particle swarm optimization was applied in the optimization models. To evaluate the effectiveness of the proposed framework, results from the optimization system as well as habitat suitability models in the natural flow and current condition were compared using various measurement indices, including the reliability index, vulnerability index, the Nash–Sutcliffe model efficiency coefficient (NSE), and root mean square error (RMSE). The case study results suggest that the reliability of water supply may be diminished under optimal release for environmental and demand considerations. Additionally, optimal release for the environment may not adequately protect downstream aquatic habitats. Therefore, in cases where the preservation of downstream habitats is a priority, dam removal may be a logical solution. Moreover, it is essential to acknowledge that the main limitation of the proposed method is its high computational complexity.

Integration of deep learning and improved multi-objective algorithm to optimize reservoir operation for balancing human and downstream ecological needs

Dam (reservoir)-induced alterations of flow and water temperature regimes can threaten downstream fish habitats and native aquatic ecosystems. Alleviating the negative environmental impacts of dam-reservoir and balancing the multiple purposes of reservoir operation have attracted wide attention. While previous studies have incorporated ecological flow requirements in reservoir operation strategies, a comprehensive analysis of trade-offs among hydropower benefits, ecological flow, and ecological water temperature demands is lacking. Hence, this study develops a multi-objective ecological scheduling model, considering total power generation, ecological flow guarantee index, and ecological water temperature guarantee index simultaneously. The model is based on an integrated multi-objective simulation-optimization (MOSO) framework which is applied to Three Gorges Reservoir. To that end, first, a hybrid long short-term memory and one-dimensional convolutional neural network (LSTM_1DCNN) model is utilized to simulate the dam discharge temperature. Then, an improved epsilon multi-objective ant colony optimization for continuous domain algorithm (ε-MOACOR) is proposed to investigate the trade-offs among the competing objectives. Results show that LSTM _1DCNN outperforms other competing models in predicting dam discharge temperature. The conflicts among economic and ecological objectives are often prominent. The proposed ε-MOACOR has potential in resolving such conflicts and has high efficiency in solving multi-objective benchmark tests as well as reservoir optimization problem. More realistic and pragmatic Pareto-optimal solutions for typical dry, normal and wet years can be generated by the MOSO framework. The ecological water temperature guarantee index objective, which should be considered in reservoir operation, can be improved as inflow discharge increases or the temporal distribution of dam discharge volume becomes more uneven.

Multiobjective multihydropower reservoir operation optimization with transformer-based deep reinforcement learning

The paper introduces a transformer-based deep reinforcement learning (T-DRL) approach designed to address the multiobjective multihydropower reservoir operation optimization (MMROO) problem. Unlike existing literature that primarily focuses on maximizing power generation from individual reservoirs, the MMROO model in this study considers the broader context of multiple reservoirs, encompassing total power generation, ecological protection, and residential area water supply. The computational challenges posed by the numerous constraints and nonlinearities of multiple reservoirs render conventional multiobjective evolutionary algorithms both expensive and lacking in generalization capabilities for solving the MMROO problem. To overcome these challenges, the paper proposes a T-DRL approach that leverages the multihead attention mechanism within the encoder module to adeptly extract complex information from reservoirs and residential areas. The two-stage encoder effectively processes diverse information separately. The multireservoir network of the decoder then generates optimal decisions based on contextual information. The case study focusing on Lake Mead and Lake Powell in the Colorado River Basin demonstrates the efficacy of the T-DRL approach, producing operation strategies that outperform a state-of-the-art method. Specifically, the proposed approach yields a 10.11% increase in electricity generation, a 39.69% reduction in amended annual proportional flow deviation, and a 4.10% rise in water supply revenue. Overall, the T-DRL approach emerges as an effective method for the multiobjective operation of multihydropower reservoir systems.

Salinity management of reservoirs by linking hydrodynamic model, surrogate model, and evolutionary optimization

This study proposes a combined system for salinity management of reservoirs in which the lake ecosystem simulation is integrated with the reservoir operation optimization. A finite volume-based depth-averaged model is applied for simulating salinity in the reservoir for a long-term period. Then, a surrogate model is developed by applying outputs of the fluid dynamic model using adaptive neuro-fuzzy inference system. The surrogate model is used in the structure of the optimization model to estimate the average salinity concentration in the reservoir. Two objectives are defined in the reservoir operation optimization including minimizing water supply loss and mitigating salinity impacts on the aquatic habitats in the lake ecosystem. According to case study results, the fluid dynamic model is reliable for simulating salinity distribution in the reservoir, which means it is recommendable for simulating salinity distribution of reservoirs. Moreover, The Nash–Sutcliff coefficient of surrogate model is 0.79, which implies it is reliable for applying in the optimization model as a surrogate model of salinity. Based on the environmental considerations, 0.55 ppt was defined as the average threshold of habitat suitability. Average optimal salinity during the simulated period is 0.52 ppt, which implies the optimization model is able to reduce salinity impacts properly. We recommend using the proposed method for the case studies in which increasing salinity is an environmental challenge for the aquatic species those living in the artificial lakes of large dams.

Multi—Objective Reservoir Operation Optimization by Considering Ecosystem Sustainability and Ecological Targets

Multi—Objective Reservoir Operation Optimization by Considering Ecosystem Sustainability and Ecological Targets

Performance Evaluation of Multilayer Perceptron Neural Network and Adaptive Neuro-Fuzzy Inference Systems for Reservoir Operation Optimization: a Case Study of Cheffia Reservoir, Algeria

Artificial Intelligence based prediction has wide applications, including hydrology, water resources management and particularly reservoir operation. Thus, two black box models based on Artificial Neural Networks and Fuzzy Logic methods are implemented and tested in forecasting reservoir operation; the first is a Multilayer Perceptron Neural Network and the second is an Adaptive Neuro-Fuzzy Inference System combining the two methods. The developed models consist of predicting evaporation, inflows and reservoir storage from their historical records and that, with aim of providing the best fit between predicted and observed values and of improving operating rules on storage and releases. The performance of achieved results demonstrated the pertinence of Artificial Neural Networks and fuzzy Logic methods in predicting cyclical state variables, such as evaporation and storage, while the prediction of inflows to reservoir generally gave better results compared to other research works available in open literature on one hand. On other hand, it is deduced from testing different prediction models that these methods are unable to predict random variables.