Performance Evaluation of the Ant Colony Optimization Algorithm for the Optimal Operation of a Multi-Reservoir System: Comparing Four Algorithms

Optimal operation of multi-reservoir systems is amongst the important problems in the field of water engineering which cannot be easily solved. Nowadays, water engineering experts are using Meta-heuristic algorithms to solve this problem due to the computational capabilities of these algorithms. Ant Colony Optimization algorithm is one of the algorithms which is based on foraging behavior of real ants. In this research, the capability of four different ant colony optimization algorithms, Ant System, Elitist Ant System, Ranked Ant system, and Max-Min Ant System has been tested to solve the four- and ten-reservoir benchmark systems. For each algorithm two different formulations are proposed; in the first formulation reservoir releases and in the second formulation reservoir storages are considered as decision variables of the problem. The results showed that the Ant Colony Optimization algorithm is a competent algorithm to solve Multi-reservoir system operation optimization problem. The Max-Min Ant System is also assessed as superior to other algorithms for such problems.