关键词: 混沌, 蚁群算法, 多峰函数, 优化, 自适应调整策略

Abstract:

In order to quickly and accurately obtain the global and local peaks of multimodal functions, this paper proposes an optimization method for multimodal functions based on the henon chaotic ant colony algorithm. In this method, the numerical characters composed by numerical solutions of complex functions are turned into the nodes of the city distribution network on the ants searching paths, the ant colonies whose number is the same as the number of function variables are constructed to solve the problem by global search, and chaos mapping technology is used to update the pheromones on the ant colony optimization path adaptively. The solving performance of this optimization method is verified by adopting low-dimensional and high-dimensional benchmark test functions and compared with those of gravitational search algorithms and other methods in literature. The comparison results indicate that, in low-dimensional multimodal function optimizations, the search efficiency of this method is two times higher than that of other methods in literature, when the number of dimensions of high-dimensional functions is greater than five, the optimization efficiency of this method is basically the same as that of other methods. However, the ability of obtaining all the global and local solutions and the solving accuracy are much higher than those of other methods.

Key words: chaos, ant colony algorithm, multimodal function, optimization, self-adaptive adjust tactics