Research Of Ant Colony Algorithms Hybrid Gentic Algorithms With Application

Ant colony algorithm is mimic the shortest path theory of the natural true ants from ant cave to food source in foraging process, and put forward the principle of a new simulation algorithm. It can carry out intelligent search and global optimization, and has the advantages of the very high parallel, robustness, collaborative and positive feedback etc.It also can good solve the complex optimization problems. It is a potential of the simulation algorithm with a number of excellent practical value which brought out in recent years. Genetic algorithm is simulation the random search algorithm of nature biological in the evolutionary process, individual fitness have improved through the role of the natural selection, heredity and variation. This algorithm has the global search ability without the relationship of the problem domain, and unfavorable into local optimal and can use evaluation function as heuristic information.Due to the ant colony algorithm in early vulnerable to suffer the lack of pheromones, causing the search time prolonged, and existing premature convergence, easily trapped into local optimal and cannot search optimal solutions in maximum range in operation process. While genetic algorithm has the ability of the global search quickly. But it is not good use of the feedback information, often leads to inaction redundancy iteration, make solving efficiency is not high. Integrated the two algorithm based on the characteristics of ant colony algorithm and genetic algorithm, and overcome the two algorithm's respective shortcomings, using the optimum combination ability of the genetic algorithm to determine the optimal parameter combination. Using clustering results obtained by ant colony algorithm, complementary strengths, optimization of this algorithm is improved, making the hybrid algorithm efficiency in convergence speed and solution on the global has been greatly improved.Based on consulting a lot of reference material at home and abroad, according to the advantages and disadvantages of two algorithms, Integrated the two algorithms and formed the hybrid algorithm strategy: in the first phase by using the global search ability of the genetic algorithm, form the initial solution quickly, after satisfy the termination conditions, transformed the dispatching optimal solution of genetic algorithm into the initial pheromone of ant colony algorithm, then by utilizing characteristics of the ant colony algorithm ,positive feedback, and efficient ,quickly form optimal solution.In this paper, the main work is after state the principle and application of the two algorithms, r presented a new hybrid algorithm mathematical model. In the solving process improvement of select strategy of the hybrid algorithm. makes the probability of the algorithm into local solution reduce, dynamic adjustment on local pheromones and global pheromone with adaptive pheromone update strategies, use current solution the biggest range. In order to evaluate the performance of hybrid algorithm, this paper simulated test the hybrid algorithm in classic optimization problem traveling salesman problem (TSP). The experimental results show that the hybrid algorithm not only accelerate the ant colony algorithm convergence speed, and improve the quality of the obtained optimal solution. Finally, according to requirements of computer application prospect of group question, proposed the idea of development sheet system, and make feasibility analysis and application.