Shrinking Pareto Fronts to Guide Reservoir Operations by Quantifying Competition Among Multiple Objectives
Hong‐Ru Wang, Fangfang Li, Guang‐Qian Wang, Jun Qiu
Abstract
Abstract Multiobjective optimization has been widely applied to reservoir operations in order to provide balanced operational schemes considering their multiple functions, including flood control, power generation, and ecological objectives. The Pareto front derived from multiobjective optimization is a set of optimal solutions that cannot quickly provide direct guidance for decision‐makers. In this study, a shrinking method is proposed to reduce the selection range of the optimal solutions on the Pareto front. Based on two proposed indices, that is, competitiveness and the competition efficiency between each pair of dual objectives, the optimal solutions are shrunk twice to accurately focus on the optimal solution region and reduce the difficulty of decision‐making. The proposed methodology is applied to a large‐scale reservoir on the upper reach of the Yellow River, China, simultaneously considering power generation, hydropower output stability, and ecological objectives. The results show that the proposed method could reduce the Pareto front to the solutions performing well in objectives and can be generalized for other multiobjective optimization models.