Abstract: The key of fuzzy controller’s design is the design of fuzzy rules with expert experience. While the expert experience can’t be obtained, the fuzzy rules should be self constructed. Ant colony algorithm is a novel simulated evolutionary algorithm. It provides a new method to solve the complex combined optimization problem. This paper takes advantage of the ant colony algorithm to optimize the fuzzy control rules, which can settle the problem of designing fuzzy controller without the expert experience. The simulation shows the favorable result and proves the feasible of the algorithm application.

Key words: ant colony algorithm, fuzzy control, fuzzy rules optimization, combined optimization

摘要： 模糊控制器设计的关键是根据专家经验确定模糊规则。然而，在专家经验难以获取的情况下将无法进行设计，这就要求模糊规则能够自动优化。模糊规则的优化过程为前件选择后件的过程，是一个组合优化问题，本文应用蚁群算法对其进行优化。蚁群算法是一种新型的模拟进化算法，已被广泛且有效的应用到求解复杂的组合优化问题中。仿真结果显示了蚁群算法应用于优化模糊规则的可行性和有效性,扩大了蚁群算法的应用范围，也为模糊控制器的设计提供了新的思路。

关键词: 蚁群算法, 模糊控制, 规则优化, 组合优化