Optimal operation of water resources systems using MOICA algorithm with reservoir hedging approach in low‐water regions*

Abstract

AbstractIn this study, simulation and optimization models are combined to apply the reservoir hedging policy. The simulation of the study basin is done by the Water Planning and Evaluation System (WEAP) model to operate the Doiraj dam reservoir located on Doiraj River. In addition, to optimize the system, the multi‐objective imperialist competitive algorithm (MOICA) model is employed so that the first goal is to maximize the percentage of supplying demands, while the second one is to reduce the violation of allowable capacities of the reservoir during the operation period. In this regard, modelling the operation of the reservoir is conducted based on the current condition for 720 months (i.e., from October 1960 to September 2019). Finally, by introducing the optimization scenario and applying the reservoir hedging policy, the optimization of the operation of the reservoir is done, and the outcomes are compared with the outputs of the reference scenario. In this study, by considering 24 decision variables including 12 hedging level variables and 12 hedging coefficient variables, the optimal answers are yielded after 1,000 iterations. The findings show that the violation of the allowable capacities did not occur in any period in the optimization scenario, while in the reference scenario, the reservoir level reached the dead level in sequent months with more water shortage, which might result in a lack of water supply in such months and serious damages to the system. Due to the application of the hedging principles in the optimization mode, the percentage of supplying the demands in the dry months has extended between 20% and 35% compared to the reference state, which demonstrates a significant reduction in the failure rate in such months compared to the reference mode.