基于非劣分类遗传算法的多目标药物提取条件优化分析应用

Multi-objective Optimize Based on Nondominated Sorting Genetic Algorithm——Uniform Multi-objective Optimization of Extraction Conditions of Drug Application

目的研究非劣分类遗传算法在均匀试验设计多目标优化中的应用。方法对微萃取五味子的均匀试验数据建立以浸膏得率、五味子醇甲、五味子总木脂素含量的子目标模型,采用遗传算法分别对其单目标优化,NSGA对其进行多目标优化,搜索最优提取条件,比较搜索结果;利用课题组编写的Matlab2009a外挂SGALAB工具箱beta5008完成遗传算法寻优。结果单目标遗传算法优化可以得到各个目标最大时的最优提取条件,NSGA进行三目标优化时,使各子目标尽可能获得最大的解,确定的最优提取条件的效果高于均匀试验中的任何一个方案。结论 NSGA搜索的Pareto非劣解是合理的,达到了较好的效果,为试验设计最优条件选择提供了合理的方法,可推广到正交试验设计、析因试验设计的最优条件选择。

Objective To study the application of multi-ob- jective optimization analysis of Nondominated Sorting Genetic Algorithm in drug extraction of uniform experimental design. Methods Using micro- extraction schisandra data in uniform design establish three-objective func- tion. Applying simple genetic algorithm and NSGA explore the optimal ex- tracting conditions to compare their optimal extracting conditions. Using SGALAB beta5008 of the Matlab2009a plug-in tool-box, achieves the ge- netic algorithm optimization. Results Single-objective genetic algorithm optimization can obtain the optimal extraction conditions of each objective. NSGA for three-objective optimization, we can get optimal extraction con- ditions,the effect of which is best. Conclusion NSGA provides reasona- ble pareto optimal solutions. It is a reasonable method for selecting optimal conditions for uniform experimental design. This method can be extended to the selection of the optimal conditions in the uniform experimental design and factorial design.