Optimized Operation Of A Multipurpose Reservoir By Evolutionary Algorithm For Panam Reservoir Project In Eastern Gujarat, India
Auteurs : Panchal S.l. . Suryanarayana T.m.v. .
Résumé
Water is life. Survival of lives with satisfying their water-based demands becomes a challenging task for water resources engineers in upcoming times. The only way to optimize potential solutions is to manage current water resources with greater attention, plan ahead and conserve them by using appropriate optimization strategies. Owing to the drawbacks of traditional optimization methods, an evolutionary algorithm-based approach inspired by nature is implemented to determine the operational strategies of challenging reservoir systems that exist in real life. Evolutionary Algorithm (EA) namely Genetic Algorithm (GA) has been developed with primary objective to establish operational strategies for Panam reservoir project, located in eastern part of Gujarat, India. In this study, the objective function is to minimize the annual sum of squared deviation from intended irrigation release and desired storage volume. The GA model was run with ten years of inflow, release, demand, surplus and evaporation data to derive an optimal reservoir operational policy. Releases from the reservoir for domestic, industrial and irrigation purposes during concern time period are the decision variables; these are subject to restrictions on the reservoir's mass balance, storage capacity, release and surplus. The results obtained shows that a minimum of 46.67 Mm3 and a maximum of 415.83 Mm3 of water may be saved in the water year 2018-19 and 2019-20, respectively. When produced results were compared to the actual releases, it emerged that GA greatly outperformed traditional optimization techniques in satisfying downstream demands and the reservoir's optimal operation may save significant quantities of water.
Mots clés
Evolutionary Algorithms (EA) ; Genetic Algorithm (GA) ; Reservoir optimization
