Codification of optimal operation policies of reservoirs in the Gorganrood basin using Wolf Search Algorithm (WSA)

In recent decades, Evolutionary algorithms have been applied successfully in various water resource engineering and management issue especially in optimal operation of reservoirs. In this research, a metaheuristic algorithm called Wolf Search Algorithm (WSA), has been developed in MATLAB software, with the purpose of optimal allocation strategies of a Multi-reservoirs system (Golestan and Voshmgir dams) located at Gorganrood Basin (North of Iran), for a five year period (from 2007-2008 to 2011-2012). At first, the performance of the developed model was investigated through several standard test functions. Next, the developed model is applied for monthly allocation of Golestan and Voshmgir reservoirs system. The results of the developed model were compared with Genetic Algorithm (GA). The objective function was defined as the “minimization of the total deficit for the study period”. In order to performance evaluation of the investigated Algorithms, two criteria of reliability (temporal and volume) and vulnerability have been used. The WSA and GA were capable to supply 95.37 and 87.07 percent of Golestan dam water demand, respectively. For the Voshmgir dam, the mentioned models could supply 93.8 and 87.59 percent of water demand, in same order. The temporal reliability (α=0.9) for WSA and GA models, was obtained 81.67 and 26.67 percent for Golestan dam and 83.33 and 38.33 percent for Voshmgir dam, respectively, revealed that the WSA was superior in optimal allocation of Multi-reservoirs system.