Deriving Optimal Operation Rule for Reservoir System Using Enhanced Optimization Algorithms

Abstract

Abstract Recently, there has been increased interest in using optimization techniques to find the optimal operation for reservoirs by applying them to various aspects of the reservoir operating system, such as finding the optimal rule curves for reservoirs. The use of different algorithms (artificial bee colony (ABC), particle swarm optimization (PSO), genetic algorithm (GA), firefly algorithm (FA), invasive weed optimization (IWO), teaching learning-based optimization (TLBO), and harmony search (HS)) was investigated in this study by integrated every algorithm to a reservoir simulation model to search for the optimal rule curves for the Mujib reservoir in Jordan from the year 2004 to 2019. To evaluate the long-term impact of water shortages and excess releases on the Mujib reservoir's operation and the effect of water demand management when reducing it by 10%, 20%, and 30% in the reservoir. The findings showed that the algorithms effectively reduced cases of water shortage and excess release compared to the current operation. The best solutions using the TLBO algorithm reduced the frequency and average of the water shortage to 55.09 % and 56.26 %, respectively, and reduced the frequency and the average of the excess release to 63.16 % and 73.31 %, respectively. The findings highlight the impact of water demand management of the reservoir on the decrease in frequency and average of the water shortage, explaining the inability of the reservoir to supply water in some months and the possibility of exposure to a shortage of water for long periods.