自适应基因表达式程序设计研究及应用

摘要

The gene expression programming is a new generic algorithm based on genome and phenomena with many GA (Genetic algorithm) and GP (Genetic programming) merits, but there are many deficiencies, specially its convergence speed is slow, it is easy to fall in local best and its fitting degree is low. An adaptive gene expression programming algbrithm (AGEP) was presented, whose chaos recombination, mutation operation and cataclysm operator were used in GEP. It was used in an application example and its results were compared with those obtained by traditional GEP and an improved GEP. The results show that the new method (AGEP) can overall convergence and it not only increases its precision and convergence speed, but also overcomes premature convergence. Improved AGEP performed better than traditional GEP and an improved GEP.%针对基因表达式程序设计(GEP)是基于基因型和表现型的新型遗传算法,它综合了遗传算法(GA)和遗传程序设计(GP)的优点,但在解决具体问题时有收敛速度较慢、易陷入局部最优和拟合度不高等缺陷,提出一种自适应基因表达式程序设计算法(AGEP),它将差分突变搜索、混沌重组和变异操作、灾变算子运用于GEP中;最后将其应用于实例中,并将其所得结果与传统的基因表达式程序设计结果进行比较.研究结果表明:该算法不仅提高了算法的精度和收敛速度,而且有效地克服了不成熟收敛,理论证明该算法全局收敛；改进的基因表达式程序设计性能良好.