Optimal multireservoir operation for flood control under constrained operational rules

Abstract

This paper presents a multireservoir flood control model that incorporates operating rules to alleviate the risk that results from traditional flood control models. The model is accurately reformulated into one that can be solved with the mixed integer linear programming (MILP) and approximated with a two‐stage linear programming (TSLP) to speed up the solution by excluding all the binary variables. A scroll decision‐making strategy is proposed by assuming only a few days of future inflows being predicted with certainty. With 10 historical and 30 designed floods in different magnitudes, the models and methods are applied to the Yangtze River basin, which includes 11 cascade reservoirs. The results on average show the MILP reduces the rule violation by 17%, while the TSLP is more than 150 times faster in CPU time. Excluding one of the reservoirs from the simulation reveals that the Three Gorges (TG) contributes the most in flood detention, accounting for 76% on average for once‐in‐a‐century floods. Interestingly, with the scroll decision‐making strategy implemented, the change from 3 to 7 days of certain forecasting has the most significant reduction in both the rule violation and flood detention, indicating that the forecasting accuracy of more recent inflows is more valuable for flood control.

Research on hybrid reservoir scheduling optimization based on improved walrus optimization algorithm with coupling adaptive ε constraint and multi-strategy optimization

“…Various techniques have been developed over the past few decades to address reservoir scheduling problems 9 .These include linear programming 9 – 11 , nonlinear programming 12 and dynamic programming 13 – 15 .However, each of these methods has its limitations. While nonlinear programming methods have their advantages, they tend to have relatively slow convergence speeds and require longer computation times, posing significant challenges in practical applications 16 , 17 .…”

Application of HSMAAOA Algorithm in Flood Control Optimal Operation of Reservoir Groups

“…There are complex hydrological and hydraulic connections among the elements of the flood control system, so the flood control operation problem of reservoir groups has the characteristics of strong constraint, multi-stage, nonlinearity, and high dimension [8]. In the past, most scholars used traditional algorithms, such as dynamic programming [9][10][11] and linear programming [12,13], to solve this problem. However, with the increase in the operation period and the number of reservoirs, the problems of slow convergence and "dimension disaster" will appear [14].…”

Application of the IGWO Algorithm for Flood Control in Reservoir Groups in Optimal Operation

“…Currently, the commonly used methods for solving flood control optimization models include traditional and intelligent optimization algorithms (Needham Jason et al 2000;Cheng et al 2008). Traditional optimization algorithms such as linear programming (Senlin Chen et al 2017;Needham Jason et al 2000;Zetai et al 2022), nonlinear programming (Unver et al 1990), and dynamic programming (Yakowitz 1982;Zhao et al 2017;Cervellera et al 2006) have been utilized. However, research has shown that as the operating cycle prolongs, the number of reservoirs increases, and the time step size increases, these methods exhibit a decrease in convergence speed and face the problem of the "curse of dimensionality" (Diao et al 2021;Shimin Li et al 2020a).…”