Bi-Level Optimization for Determining Operating Strategies for Inter-Basin Water Transfer-Supply Reservoirs

Abstract

Operating strategies for inter-basin water transfer-supply reservoirs play an important role in assisting in the decision-making process of effectively transferring and supplying water. This article presents a bi-level optimization model for determining operating strategies for inter-basin water transfer-supply reservoirs. The bi-level model avoids the structural description limitation of traditional optimization models, which normally optimize all of the variables simultaneously, by exactly describing the leader-follower framework used in water transfer-supply. The bi-level optimization problem was solved by using Adaptive Genetic Algorithm (AGA) and was compared with the traditional optimization model. Furthermore, to improve the solution procedure, two methods were used to improve the initial solution of the bi-level model, both based on probability description methods and chaos optimization. The method was applied in a case study of Biliu River reservoir, which is located in Liaoning Province, Northeast China. The results demonstrated that the bi-level model mathematically represents the hierarchical structure of inter-basin water transfer-supply operating strategies, but it requires more time than traditional models to find optimal solutions. Thus, this article proposes methods to improve the solution procedure. The results showed that the proposed bi-level model with the improved solution procedures achieves better reservoir operation performance and shorter calculation times than the general solution procedure when the same decision variables are considered. The amounts of transferred water and abandoned water decreased by 3.8% ~ 9.3% and 5.4% ~ 12.1%, respectively, thus representing an improvement in the efficiency of water transfer. Moreover, the calculation time was decreased by approximately 60%, relative to that of the general solution procedure.