Optimal Operation of Reservoirs Considering Water Quantity and Quality Aspects: A Systematic State-of-the-Art Review

Researchers have been placing greater emphasis on improving the operating policies of multi-purpose reservoirs in recent years. This emphasis has addressed various issues, including water supply to demands, energy production, water quality management, and flood risk reduction. This paper uses the meta-analysis method (PRISMA) for article selection and systematic review of papers focusing on water quality management in reservoirs using simulation-optimization models. After conducting a comprehensive search, 42 papers were selected for in-depth review. This paper offers an in-depth examination of various coupled simulation and optimization models used for optimal reservoir operation, focusing on addressing water quality concerns. The findings reveal a significant preference among researchers for using genetic algorithms (GA) (55%) and multi-objective particle swarm optimization (MOPSO) (30%) for optimizing reservoir operating policies. Moreover, the analysis highlights that most (81%) of the reservoir water quality simulation models employed in these papers are one-dimensional or surrogate models. These simulation models are favored for their shorter simulation time compared to two-dimensional models. The review identifies existing gaps and proposes future directions for advancing this field. For instance, there is an opportunity to enhance both inflow prediction accuracy and reservoir water quality predictions by exploring diverse data obtained from remote sensing, unmanned aerial vehicles (drones), unmanned underwater vehicles, and numerical weather prediction models. The obtained data from different sources can be fused using artificial intelligence techniques to improve their accuracy.