模糊C均值算法因其简单、快速得到了广泛应用，但仍存在对初始值敏感和容易陷入局部最优的不足。提出了一种新的小生境萤火虫模糊聚类算法。该算法使用遍历性较好的立方混沌映射序列初始化萤火虫种群，并将随机惯性权重引入萤火虫算法，改变了基本萤火虫算法的位置更新公式，不仅减少了迭代次数，而且平衡了算法局部搜索和全局搜索的能力；并在迭代过程中合适时机实施小生境算法，进而增加了种群的多样性并加快了算法运算速度。仿真实验结果表明，该算法有效地抑制了早熟，并保证了种群的多样性和避免陷入局部最优，取得了较好的稳定性及良好的聚类结果。

The fuzzy C-means algorithm is widely used due to its simplicity and speediness. However, it is sensitive to  the initial value and easy to fall into local optimum. We propose a new fuzzy clustering based on niching firefly. The algorithm utilizes the chaotic sequence to initialize the firefly population so as to obtain the initial population. The introduction of random inertia weight not only decreases the number of iterations, but also balances the global search ability and the local search ability of the algorithm. By implementing the niche in the process of the iteration algorithm, the diversity of population is increased and the algorithm’s speed is accelerated. Simulation results show that the proposed algorithm can suppress precociousness effectively and ensure population diversity. It can also avoid falling into the local optimum and achieve good clustering performance.
相关文章