Optimization of selective withdrawal systems in hydropower reservoir considering water quality and quantity aspects
Introduction
Water reservoirs are the main elements of water supply systems that serve multiple functions, including municipal water supply, hydropower generation, irrigation, and navigation. Nowadays, many reservoirs worldwide are suffering from excess nutrient inflows and water quality deterioration. These human-made water infrastructures affect the river ecological environment in many ways, including changes in the cycles and rhythms of river natural flow and alteration of thermal and water quality patterns in the water bodies. The aforementioned processes, have generally been threatening water security on both local and global scales. Under the risk of water scarcity, appropriate design and operation practices emphasizing the environmental aspects are crucial in the efficient management of reservoirs as valuable freshwater resources. These new and modern reservoir management practices include a new paradigm which tries to balance economic and social objectives against water quality conservation requirements.
Water reservoirs could be equipped with selective withdrawal structures, consisting of multiple intakes drawing water from multiple depths. In this technique, multiple intakes could be opened simultaneously to mix water from various elevations to help achieve in-reservoir and downstream temperature and other water quality objectives. The design of SWS (vertical locations and withdrawal ratio) and adjustment of withdrawal patterns have been known as new alternatives to enhance in-reservoir and downstream river water quality. Although the application of SWS could help fulfill desirable water quality conditions in water bodies, it could negatively affect the energy generation objectives in reservoirs with hydropower objectives. Consequently, the design and operation of SWSs in reservoirs should be studied in integrated water resources management with quality and quantity aspects in mind. Saito et al. (2001) studied the effects of installing a temperature control device on phytoplankton production in a dam at Shasta Lake, California. CE-QUAL-W2 model was applied to simulate and predict the impacts of new operation strategies on the lake water quality responses. Gelda and Effler (2007) developed a simulation model framework to evaluate the benefits of multi-level intake configurations to improve reservoir outflow temperature and turbidity. A 2D hydrodynamic and water quality simulation model (WQSM), CE-QUAL-W2, was coupled with a heuristic search algorithm to support the automated determination of intake levels through evaluating the efficacy of various combinations of the number and position of intake. The proposed methodology was applied in Schoharie Reservoir, New York, to enhance the downstream temperature and water quality under the conditions of a critical historic year. The solutions serving as the best intake levels to alleviate the adverse effects of reservoir thermal stratification and turbidity issues do not necessarily guarantee the overall optimum due to the short-time frame in planning. In other words, the long-term hydrological and meteorological variability and different initial storage conditions have to be considered in an optimal design of SWSs. Çalışkan and Elçi (2009) examined the effects of SWS to control released water temperature. A 3D hydrodynamic model (EFDC) for Tahtali Reservoir in Turkey was calibrated and validated to evaluate the effects of SWSs on water mixing and anoxia reduction in the reservoir. Rheinheimer et al. (2015) developed a mathematical linear programming model to determine the water released from different thermal pools in Lake Spaulding. The developed model aimed to mitigate the deviations from target downstream temperatures in South Fork Yuba River in California with climate warming. A dynamic thermal model with simplified assumptions was applied to simulate downstream water temperature. Rounds and Buccola (2015) modified the algorithm in the CE-QUAL-W2 model (3.7) for blending the outflow leaving multiple reservoir outlets (up to 10 outlets) to meet the outflow target temperature on a time series scale. The revised version of the CE-QUAL-W2 model showed the capability to simulate the effects of floating outlets in conjunction with lower fixed-elevation outlets ranking the priority designations and criteria for head and flows into account. The new model features were implemented in Detroit Lake under various environmental conditions. Li and Qiu (2015) developed a simulation–optimization (SO) model to maximize hydropower energy generation and sediment outflow fluxes in Three Gorges Reservoir (TGR), China. In order to manage the mismatch between increased power generation and enlarged sediment scouring, a multi-objective optimization framework was applied to balance the water and sediment content in the reservoir. The TGR was equipped with turbine intake, upper and lower spillways used for power generation, and sediment fluxes. Soleimani et al. (2016) extracted optimal operating rules in selective withdrawal systems considering downstream thermal pollutions as the only objective. LIBSVM model as the surrogate model of CE-QUAL-W2 was coupled with a genetic algorithm to derive monthly operating rules in the selective withdrawal scheme to mitigate reservoir outflow thermal pollutions. He et al. (2017) studied the effects of the temperature-control curtain (TCC) as an effective selective withdrawal facility on both the reservoir outflow temperature and the thermal structure in Sanbanxi Reservoir. CE-QUAL-W2 model of Sanbanxi Reservoir was calibrated to simulate the flow and temperature under different TCC scenarios. Weber et al. (2017) developed a sustainable management model of drinking water reservoir considering water supply and environmental impacts. They derived an optimal withdrawal strategy with the aim of achieving the desirable downstream water temperature and avoiding unacceptable low oxygen concentration within the reservoir. GLM-AED2 model was calibrated and validated to evaluate various withdrawal strategies in Grosse Dhuenn Reservoir to select the optimal scenario. Yu et al. (2018) developed optimal reservoir operation strategies in the TGR system to manage the eutrophication challenges in downstream estuarine and maximize hydropower energy generation. The effects of TCC on reservoir water quality and ecology were studied by He et al. (2019). They considered multiple typical water-retaining proportions (the ratio of the water-retaining area of the TCC to the cross-sectional area of the TCC) and future climatic changes to analyze water quality and ecology. Kim and Choi (2021) studied the effects of the selective withdrawal framework (SWF) on reservoir outflow water temperature and downstream fish habitat in Soyang-gang Dam, Korea. The physical habitat simulations were carried out using habitat suitability curves (HSCs) for adult and spawning Zacco platypus in the study area. The results indicated that using selective withdrawal operations, the effects of water impounding on fish habitat could be reduced. Duka et al. (2021) studied the effects of SWF and upstream reach control through the vertical curtain on the thermal stratification responses of reservoir with 3D WQSM.
As highlighted, most of the recent researches on SWS design and reservoir operation studies has used lumped WQSMs with simplifying assumptions (Rheinheimer et al., 2015, Li and Qiu, 2015). Disregarding the long-term hydrological and meteorological variabilities and the sensitivity to the initial conditions (Gelda and Effler, 2007, Li and Qiu, 2015, Yu et al., 2018), they have either identified the best solutions according to trial and error methods (Saito et al., 2001, Çalışkan and Elçi, 2009, Rounds and Buccola, 2015, He et al., 2017, He et al., 2019, Weber et al., 2017, Duka et al., 2021, Kim and Choi, 2021), and/or focused solely on quality objectives (Gelda and Effler, 2007, Soleimani et al., 2016). Very limited studies on SWSs optimization (Saadatpour et al., 2020) have considered hydropower energy generation and water quality issues under an extensive hydrological period featuring wide ranges of reservoir drawdown and refill. Also, limited researches have focused on SWS design and /or have embedded parametric reservoir operation rules focusing on both quality and quantity aspects in a selective withdrawal scheme. This research attempts to overcome the aforementioned limitations.
In specific, this study introduces an integrated modeling framework for an optimal SWS design followed by the derivation of the optimal operation rules emphasizing the selective withdrawal framework to increase hydropower peak energy generation and enhance in-reservoir water quality and downstream temperature. This study intends to integrate and resolve the challenges associated with applying metaheuristic algorithms and analytical tools, spatial and temporal level of potential water quality and quantity impacts, cognitive complexity of 2D hydrodynamic and WQSM, and adaptive modifications of surrogate models in the emulation processes of waterbody’s responses. The paper is organized as follows. Section 2 elaborates the case study and research problems. Section 3 introduces the applied algorithm and analytical tools, examines how the problem definitions affect the decision vectors, and presents the problem formulations. Section 4 discusses the results and Section 5 presents the concluding remarks of this study.
Section snippets
The study area
Karkheh reservoir, as a vital water supply system of a semi-arid region, is located in southwestern Iran. The construction of the Karkheh reservoir was completed in 1998 by Iran Water and Power Company. With 5 × 109 m3 capacity, 64 km length, and 162 km2 surface area at normal water levels (220 masl), this water body is the largest artificial lake in Iran. Karkheh Dam is equipped with three outlets located in vertical positions in the impoundment. The bottom outlet is used for sediment
Problem statement
In the sustainable operation of hydropower reservoirs, there is a desire to maximize hydropower energy generation whilst alleviating the downstream temperature alterations and enhancing in-reservoir water quality. These aims are studied by modifying reservoir operations in the selective withdrawal framework. The preferable operations of water reservoirs involve effective design and management of water infrastructures by integrating simulation and optimization tools. The problem is that the
Surrogate model development
The learning data-set to adjust surrogate model (Ŝ) was developed by running high fidelity hydrodynamic and WQSM (calibrated and validated CE-QUAL-W2 model of Karkheh reservoir (Afshar & Saadatpour, 2009)) over the time horizon 2000–2016. Various withdrawal schemes (30 withdrawal scenarios) were generated randomly to construct the reduced state-decision space as uniformly as possible. The input data selections of the surrogate model could be defined as the task of selecting a proper subset of
Conclusion
In this paper, ASBSOF was developed based on the adaptive and recursive attitude to a) arrange SWSs in reservoir optimally and b) derive selective withdrawal rules considering quality and quantity objectives. The proposed methodology tried both to satisfy hydropower energy generation and to moderate in-reservoir DO anaerobic conditions and downstream temperature pollutions. Considering the requirements of comprehensive hydrodynamics and WQSM, including selective withdrawal algorithm,
CRediT authorship contribution statement
Motahareh Saadatpour: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing. Shima Javaheri: Data curation, Formal analysis, Methodology, Resources, Software, Visualization, Writing - original draft. Abbas Afshar: Investigation, Methodology, Supervision, Validation, Visualization, Writing - review & editing. Samuel Sandoval Solis:
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
References (38)
- et al.
Multi-level information fusion for spatiotemporal monitoring in water distribu tion networks
Expert Systems with Applications
(2015) - et al.
A surrogate-assisted evolution strategy for constrained multi-objective optimization
Expert Systems with Applications
(2016) - et al.
Thermal stratification responses of a monomictic reservoir under different seasons and operation schemes
Science of The Total Environment
(2021) - et al.
A surrogate-assisted multi-objective particle swarm optimization of expensive constrained combinatorial optimization problems
Knowledge-Based Systems
(2021) - et al.
Modeling the effect of temperature-control curtain on the thermal structure in a deep stratified reservoir
Journal of Environmental Management
(2017) - et al.
Assessment of the impact of selective withdrawal on downstream fish habitats using a coupled hydrodynamic and habitat modeling
Journal of Hydrology
(2021) - et al.
A modified particle swarm optimization using adaptive strategy
Expert Systems with Applications
(2020) - et al.
Methods used for the development of neural networks for the prediction of water resource variables in river systems: Current status and future directions
Environmental Modelling and Software
(2010) - et al.
Multi-Objective Particle Swarm Optimization with time variant inertia and acceleration coefficients
Information Sciences
(2007) - et al.
An improved multi-objective particle swarm optimizer for multi-objective problems
Expert Systems with Applications
(2010)
Surrogate-enhanced evolutionary annealing simplex algorithm for effective and efficient optimization of water resources problems on a budget
Environmental Modelling & Software
Optimizing withdrawal from drinking water reservoirs to reduce downstream temperature pollution and reservoir Hypoxia
Journal of Environmental Management
Protocol for developing ANN models and its application to the assessment of the quality of the ANN model development process in drinking water quality modelling
Environmental Modelling and Software.
Optimal reservoir operation using multi-objective evolutionary algorithms for potential estuarine eutrophication control
Journal of Environmental Management
Eutrophication in reservoirs: 2D modelling of the Karkheh reservoir
Journal of Water & Wastewater
Development of a complex system dynamic eutrophication model: Application to Karkheh reservoir
Environmental Engineering Science
Effects of selective withdrawal on hydrodynamics of a stratified reservoir
Water Resources Management
Handling multiple objectives with particle swarm optimization
IEEE Transactions on Evolutionary Computation
Cited by (26)
Mitigating adverse impacts of reservoir impoundment on lake ecology: A case study of the Three Gorges Reservoir and Dongting Lake
2024, Journal of Cleaner ProductionAdaptive reservoir operation considering water quantity and quality objectives: Application of parallel cellular automata and sub-seasonal streamflow forecasts
2024, Journal of Environmental ManagementHypolimnetic aeration optimization based on reservoir thermal stratification simulation
2023, Journal of Hydrology