Abstract: To solve the problems in the cultural algorithm has a number of problems, for example, lack of a region concept in the population space, lack of cultural evolution in the belief space, and defects such as low accuracy and being easy to fall into a local optimum. This paper proposes a new cultural algorithm based on cellular space structure. It can simulate a double-layer evolutionary system of culture algorithm by embedding a cellular space grid structure in the framework of computing population space and belief space. For the population space, the evolutionary individuals are distributed in the lower cellular space grid. The grid is divided into many areas so that individuals in each area evolve independently by using differential evolution algorithm. For the belief space, the evolution information is put into the upper grid corresponding to the population space, and the evolution of culture is realized using a diffusion mechanism of culture. Experimental results show that the algorithm is effective in convergence accuracy and global search capability, and has advantages in dealing with complex high-dimensional optimization problems.

Key words: cultural algorithm, cellular space structure, differential evolution, complex high-dimensional optimization problem