A hybrid constrained coral reefs optimization algorithm with machine learning for optimizing multi-reservoir systems operation

Author(s): Emami, Mohammad; Nazif, Sara; Mousavi, Sayed-Farhad; Karami, Hojat; Daccache, Andre | Abstract: The continuous growing demand for water, prolonged periods of drought, and climatic uncertainties attributed mainly to climate change mean surface water reservoirs more than ever need be managed efficiently. Several optimization algorithms have been developed optimize multi-reservoir systems operation, mostly during severe dry/wet seasons, mitigate extreme-events consequences. Yet, convergence speed, presence local optimums, calculation-cost efficiency are challenging while looking the global optimum. In this paper, problem finding an efficient optimal operation policy in is discussed. complexity long-term operating rules reservoirs’ upstream downstream joint-demands projected recursive constraints make formidable. original Coral Reefs Optimization (CRO) algorithm, which a meta-heuristic evolutionary two modified versions used solve problem. Proposed modifications reduce calculation cost by narrowing search space called constrained-CCRO adjusting reproduction operators with reinforcement learning approach, namely Q-Learning method (i.e., CCRO-QL algorithm). optimum solution feasible region instead entire domain. models’ performance has evaluated solving five mathematical benchmark problems well-known four-reservoir system (CFr) Obtained results compared those literature optimum, Linear Programming (LP) achieves. shown very calculation-cost-effective locating or near-optimal solutions terms convergence, accuracy, robustness.