Optimum Operation of Reservoir Using Two Evolutionary Algorithms: Imperialist Competitive Algorithm (ICA) and Cuckoo Optimization Algorithm (COA)

Abstract

Water scarcity is one of the most serious problems in many parts of the world that affects negatively on the environment, society, and economy. In order to mitigate the negative effects of this issue, optimal water resource management is pivotal. In current paper, imperialist competitive algorithm (ICA) and cuckoo optimization algorithm (COA) which they are two new evolutionary methods, were used in optimal operation of reservoir. Firstly, these algorithms were used in solving several benchmark problems. Afterwards, optimal operation policy of Karun4 reservoir was extracted. Karun4 is located in Chaharmahal Va Bakhtiari province in western of Iran. Finally, the results which obtained from these methods were compared with genetic algorithm (GA) and nonlinear programing (NLP). In benchmark problems, COA converges to optimal point appropriately well and shows best performance. In these problems, ICA represent suitable ability to achieve global optimum. Both COA and ICA algorithms showed high performance in extraction of optimal operation policies from Karun4, which was conducted over a period of 360 months, with the aim of maximizing productivity. COA indicated the best performance with average value of 5.454 for objective function, and ICA with 6.461 value was at the second rank. In addition, GA objective function value was 6.869. Also NLP solver of Lingo11 was used in order to optimal operation of Karun4 reservoir for evaluating the ability of these algorithms to achieve global solution. Objective function value which was gained by NLP method was 5.243. The results reflect the strength of COA in approaching global optimum.