Abstract:Leading raking selection strategy into basic ant colony algorithm and using a definite linear or nonlinear relation,the pheromone differences are changed into raking,and the selection probability of individual is controlled by changing selection pressure,such a selection rule possesses good global optimization performance and can inhibit premature convergence effectively. To improve the local searching performance of the algorithm,before global updating of pheromone the chaotic searching method is performed after each time of iteration to search out better solution near current optimal ants. Simulation results of IEEE 14-bus,IEEE 30-bus and IEEE 57-bus system validate the effectiveness of the proposed algorithm. In addition,the action of chaotic search to algorithm stability is pointed out.

Ant Colony Algorithm Based on Chaos Theory and Ranking Selection in Reactive Power Optimization