基于PARETO的改进遗传在多目标模型的研究

The Research on the Multi-objective Case Using Improved GA Based on PARETO

该文先介绍PARETO解及多目标的相关概念,再通过自适应更新机制、精英保留策略等方法来提高遗传搜索效率,并且对多目标函数的结构进行改进设计,结合IAGAMO模型,以全局搜索机制作为研究基础,针对遗传算法实际应用缺陷进行了分析,着重论述通过全局搜索机制对提高局部搜索中遗传算法的影响,从而加速了IAGAMO混合算法的运算速度以及收敛效率。最后将PARETO的IAGAMO算法在生产实例进行仿真验证,结果所获得PARETO解的数据较符合生产的实际应用。因此,PARETO以其巨大的技术优势,有效提升了搜索效率,在多目标搜索以及解集的优化中发挥了重要的作用,因此具有广阔的发展空间。

This article first introduces the concepts of PARETO solutions and multi-target,and then through an adaptive update mechanism, elitist and other methods to improve the efficiency of genetic search, and the structure of multi-objective function to improve the design, combined with IAGAMO model, through the adaptive cross, features of the PARETO optimal solution, variability update mechanism and mixing in elite global search mechanism strategy to further address the lack of genetic algorithms to search for local solutions, and it to improve the IAGAMO hybrid convergence speed and computational efficiency. Finally, PARETO of IAGAMO algorithm simulation instance in the production, the results obtained are compared with data PARETO solution compatible with the application of production.Therefore,PARETO not only to better improve the efficiency of multi-objective optimization search solution set, but also with a wide range of produce promotional value.