Hybrid Swarm Intelligence Optimization Algorithms Based On Cultural Evolution And Their Applications

Swarm intelligence optimization algorithm is a new type of optimization algorithm to simulate the biological intelligence behavior and rule. This algorithm is inspired by the bionics. The new idea has aroused wide concern in artificial intelligence academics school. Study of existing swarm intelligence optimization algorithm is still in the exploratory stage. Thus, the performance and application in many ways still falls far short of what people imagine the level of intelligence. With the rapid development and need of current computer science, the study and improvement to the algorithm is the new way to solve the problem.Cultural algorithm is a computing framework with unique double evolutionary characteristics. The main structures include the belief space and population space. The double structures of cultural algorithm have good parallelism and strong convergence. The use of cultural algorithm can improve the performance of swarm intelligence algorithm, and overcome the inherent shortcomings in traditional algorithms. The operating mechanism of algorithm tends to mature by improving the algorithm structure and performance, and can get better results in optimization problems.In this paper, swarm intelligence algorithm is the main line and to use cultural algorithm as framework. Using the hybrid ideas improves and integrates the algorithm, and the hybrid algorithm is applied to the function optimization and constrained optimization of engineering problems. First of all, the paper introduces the basic theory of swarm intelligence optimization algorithm and the origins of cultural algorithm framework. Then the population space in the cultural algorithm use artificial fish swarm algorithm as the main part to optimize knowledge from the population. The knowledge is part of the initial solution in the belief space. The two spaces form the double evolutionary parallel state, thereby increasing the population diversity and enhancing the accuracy and convergence speed of the algorithm. Second, the belief space of cultural algorithm is improved and simulating annealing algorithm is introduced within the framework of cultural algorithm. To overcome the shortcoming of cultural-based particle swarm optimization that it is easy to trap into local minimum, the simulating annealing algorithm embedded in the cultural algorithm framework as an evolving course from the knowledge space, which respectively has its own population to evolve independently and parallel. The mechanism improves the algorithm search ability and convergence speed. Finally, the use of the chaotic search strategy improves the existing cultural particle swarm algorithm. For advantages of the algorithm, it is applied to the classic constrained optimization problems.By comparison experimental result, cultural algorithm and swarm intelligence optimization algorithm to achieve the desired integration of good results. And it can be found that this hybrid algorithm illustrate its higher searching efficiency and better stability.