Abstract: Fractal Algorithm（FA） is a kind of global optimization algorithm. It completes the search process based on the feasible region nested partition. Because of its depth first search strategy, it’s easy to  getting into the local optimum, and the convergence speed is very slow. In order to overcome the shortcomings of the original algorithm, the article puts forward an improved Fractal Optimization Algorithm（FOA）. The algorithm avoids segmenting the feasible region blind, and guides the candidate points being selected in the best area with a large probability by imitating fractal growth process. Finally, numerical experiments are done to verify the effectiveness of the algorithm, and the result shows that it has good convergence.

Key words: fractal, nested partition, fractal growth, fractal optimization

摘要： 分形算法是一种全局优化算法，通过对可行域的嵌套分割来完成搜寻过程，由于其深度优先的搜索策略，开始容易陷入局部最优，收敛速度慢。针对原算法的缺点，提出一种改进的分形优化算法。该算法模拟分形生长的过程，避免了对可行域的盲目分割，并指导新点以较大的概率产生于较优的区域。通过数值试验来验证该算法的有效性，结果表明该算法具有良好的收敛性。

关键词: 分形, 嵌套分割, 分形生长, 分形优化