求解函数优化问题的交谊舞算法

针对函数优化问题,借鉴交谊舞的有关机理,构造了若干舞蹈算子（空间移动,局部舞伴交换和概率舞池修正）.将优化过程视为在搜索空间中从候选解向最优解的信息交换过程,提出一种新的优化算法-交谊舞算法.着重研究了算法的原理机制,流程实现和改进方法.通过对复杂函数优化问题的仿真试验,结合与简单遗传算法对比分析,验证新算法具有有效防止陷入局部极小值,以及收敛速度快等性质.交谊舞算法具有解决函数优化问题的潜力.

An Algorithm Based on Social Dance Phenomenon for Function Optimization

Inspired by the mechanism of social dance phenomenon, some dance optimization operators, mainly including the partner change operator, position move operator and modify dance pool operator, were constructed function optimize. The optimization was considered as a process where information transmits from candidate solution to optimal solution in search space, and a new optimization algorithm, social dance optimization algorithm (SDOA) is explored on the basis of the effective partner exchange mechanism of the social dance phenomenon. The principle, implement and improvement way have been discussed. Compared with the genetic algorithms, the simulated results of some complex functions optimization indicate t hat SDOA enables to enhance t he diversity of the population with a higher convergence rate and avoid local voltage and genetic algorithm deceptive problem to some extent.