灾变机制下元胞遗传算法的选择压力研究

Selection pressure study of cellular genetic with disturbances

对于遗传算法,全局探索和局部寻优之间的关系是算法好坏的核心问题,选择压力的变化直接影响着两者之间的平衡。研究了元胞遗传算法的选择压力,并在其灾变机制下进一步研究了其选择压力的变化规律,通过改变灾变规模和周期观察选择压的变化。灾变的发生使整个算法的选择压力降低,当灾变规模越大、周期越短,算法的选择压力也就越小。对于不同优化问题选择一个合适的选择压能使全局探索和局部寻优之间的平衡达到最佳化从而能够又快又精确地寻求到全局最优解。

With regard to Genetic Algorithm（GA）,the exploration/exploitation trade-off is the key issue of the algorithm.The change of selection pressure affects their balance directly.This paper studies the selection pressure of cellular genetic algorithms and it also studies the selection pressure with disturbance,and observing the changes of selection pressure by changing the scale and cycle of disturbance.The disturbance weaken the selection pressure of the algorithm.When the disturbance has larger scale and shorter cycle,the algorithm has smaller selection pressure.Using an appropriate selection pressure enable the exploration/exploitation trade-off to achieve the best balance in order to find the global optimal solution quickly and accurately for different optimization problems.