Ant Colony Algorithm And Its Application In The Continuous Space Optimization

The ant colony algorithm is a kind of stochastic explorative algorithms. As the same of other kind of stimulated evolutionary algorithms, it finds the best solution of optimization problem by making uses of the evolutionary procedure of a set of cooperating agents of candidate solutions. It has showed a great deal of salient character and performed great value in its application. Choosing the analysis of character of Ant System (the basic algorithms of ant colony) as the research background, this paper focuses on the model, the behavior, the characteristics and its improvements, and raises a model of ant colony algorithm on general function optimization problems. Finally we discuss the salient character of ant colony algorithm on the continuous space optimization problems, and propose the region of further investigation.Firstly, the biological mechanism, the development and the character of the ant colony algorithm are especially discussed.The current situations of research and application of the ant colony algorithm is introduced. The shortages in solving the continuous space optimization problems are presented;The principle, the model, the characteristics and the management about the basic algorithms of ant colony are also presented.And a list of improvements of continuous space optimization have been referenced and presented.Secondly, because the slow convergence speed constrains the ACA's application to many problems, a series of improvement to the basic ACA is presented in this paper. The improved ACA is used to optimize the test and some effect is obtained in improving convergence speed.However, because of the limitation of the improved ACA, Logistic mapping and Ulam-von Neumann mapping are analyzed systematically in this paper. The intrinsic stochastic property, ergodicity, regularity and sensitivity to initial values are shown in this paper. Then a hybrid algorithm of chaos and ant colony is presented, that is Chaos- Ant Colony Algorithm. The optimum solution of test function shows that the hybrid algorithm has advantages of high precision, fast convergence and stabilization.Finally, a feasible way is presented for applying the algorithms to practice. The result about the emulated test demonstrate that the characteristics of ant colony algorithm for general function optimization problem are salient, and it is worthy to make further researches; To the problem of parameter's optimization design of PID controller, a comparative analysis of the optimization results of the ant colony algorithm is made to the Z-N algorithm. As a novel simulated evolutionary optimization method, the simulation results show that the ant colony algorithm is valid and practical.The ant colony algorithm is a kind of stochastic explorative algorithms which has showed many excellent characteristics. It is demonstrated that the ant colony algorithm is more adaptivable to the chaos algorithms. Although some successful applications have been presented, it also has many problems for solving and making further investigation.