摘要
目的 探讨多因素、多水平试验中最优试验条件的确定方法。方法 以二次响应面回归模型为目标函数 ,用遗传算法搜索最优试验条件。结果 在试验范围内 ,遗传算法确定的最优试验条件精度较高 ,且可预测响应变量的水平 ;若根据实际的可行性 ,外延因素的搜索范围 ,为研究提供新的试验方向。结论 模型与遗传算法结合 ,为多因素、多水平试验中最优试验条件确定提供了精度高 ,信息量大的新方法。
Objective Exploring a method to determine the best experimental condition of the response variable for multiple levels of multiple factors experiment.Methods On the basis of target function——the quadratic response surface regression model , the best experimental condition was searched by genetic algorithm (GA).Results The best experimental condition by GA was more accuracy, while GA could predict the value of the response variable. If extending the searching scope, according to practice, GA could clue to the new experimental direction.Conclusion The response surface regression model and GA combined to provide a precise,new method for determining the best experimental condition of the response variable for multiple levels of multiple factors experiment.
出处
《中国卫生统计》
CSCD
北大核心
2004年第4期194-197,共4页
Chinese Journal of Health Statistics
关键词
二次响应面回归模型
遗传算法
最优试验条件
GA
表达方式
Quadratic response surface regression model, Genetic algorithm(GA), Best experimental condition
