The application of harmony search and genetic algorithms for the simultaneous optimization of integrated reservoir–FARM systems (IRFS)^*

Abstract

Optimal irrigation allocation and reservoir operation are essential in combating water scarcity in arid and semi-arid regions like Iran. Due to the huge number of decision variables in a reservoir-farm system, the inseparable nature of the crop yield function and the large variety of constraints, a genetic algorithm (GA) and harmony search (HS) are employed in the current paper to construct an integrated reservoir-farm system (IRFS). This integrated model can take into account crop sensitivity to water stress in order to maximize the net benefit of crops and to optimize crop areas and irrigation scheduling. This methodology is applied to the multi-crop Aharchay irrigation system downstream of Sattarkhan Dam. The outputs include the optimal values of the cultivated area, irrigation depth, and water release from the reservoir in 10-day periods. The results showed the speed of convergence, the optimal value of the total benefit and the slightly greater yield values of the HS algorithm compared to those of the GA. Both models also resulted in higher benefit amounts compared to the real and attainable benefit in the region. Finally, regarding the speed of convergence to the maximum value of the objective function, the HS algorithm demonstrated more promising results than the GA in the current case of interest.