An adaptive particle swarm optimization algorithm for reservoir operation optimization

• ROO problem consists of many conflictive objectives to be optimized synchronously using ɛ-constraint methods.
• An improved adaptive particle swarm optimization algorithm (IAPSO) is presented.
• New output constraints handling strategy for ROO problem are proposed.
• Better quality solutions satisfying output constraint in both robustness and accuracy are obtained for ROO.

Reservoir operation optimization (ROO) is a complicated dynamically constrained nonlinear problem that is important in the context of reservoir system operation. In this study, improved adaptive particle swarm optimization (IAPSO) is proposed to solve the problem, which involves many conflicting objectives and constraints. The proposed algorithm takes particle swarm optimization (PSO) as the main evolution method. To overcome the premature convergence of PSO, adjusting dynamically the two sensitive parameters of PSO guides the evolution direction of each particle in the evolution process. In the IAPSO method, an adaptive dynamic parameter control mechanism is applied to determine parameter settings. Moreover, a new strategy is proposed to handle the reservoir output constraint of ROO problem. Finally, the feasibility and effectiveness of the proposed IAPSO algorithm are validated by the Three Gorges Project (TGP) with 42.23 bkW power generation and XiLuoDo Project (XLDP) with 30.10 bkW. Compared with other methods, the IAPSO provides a better operational result with greater effectiveness and robustness, and appears to be better in terms of power generation benefit and convergence performance. Meanwhile, the optimal results could meet output constraint at each interval.