基于遗传算法的青海土族和藏族人发元素非线性映射研究  

Study of Non-linear Mapping of Elemental Concentrations in the Human Hair Covering Tu and Zang Nationalities of Qinghai Based on Genetic Algorithms

作　　者：李仲[1] 龙启萍[2] 吉守祥[2] 

机构地区：[1]青海民族大学物理与电子信息工程学院,青海西宁810007 [2]青海民族大学化学与生命科学学院,青海西宁810007

出　　处：《榆林学院学报》2016年第6期46-49,共4页Journal of Yulin University

基　　金：国家自然科学基金资助项目(81160554);国家教育部春晖计划项目(Z2012108)

摘　　要：利用非线性映射与遗传算法相结合的方法对青海少数民族进行识别。遗传算法是一种借鉴生物界自然选择和进化机制发展起来的高度并行、随机、自适应搜索算法,由于其具有健壮性,特别适合于处理传统搜索算法解决不好的复杂的和非线性问题。它是解决各类函数优化和非线性映射问题的一种有效算法。以青海少数民族土族和藏族人发中微量元素的模式识别为例,编制基于误差函数的目标函数和遗传算法约束参数,应用基于MATLAB的遗传算法工具箱进行优化求解。结果表明:该工具箱在求解此类非线性优化问题上是有效的,基于遗传算法的非线性映射的预测结果比线性判别方法要好些。The nonlinearity mapping combined with genetic algorithms is used in the recognition of ethnic minorities of Qinghai province. Genetic Algorithm （GA） is a highly effective paralleling, random and self- applicable algo- rithm based on the development of the natural evolution and option. Due to its strength, it is particularly applicable to deal with traditional searching algorithm and tackle complicated nonlinear problems. Genetic Algorithms are ef- fective for solving all kinds of optimization problems and non - linearity mapping. By compiling object function based on error function and constraint parameters, a model of projection pursuit on pattern recognition of elemental concentrations in the human hair covering Tu and Zang nationalities of Qinghai province is solved based on MAT- LAB genetic algorithm Toolbox. Result indicates that the toolbox is effective. It has been found that the predicted result by non - linearity mapping based on genetic algorithms is better than that of linear discrimination method.

关 键 词：遗传算法 非线性映射 模式识别 土族 藏族 

分 类 号：TH112[机械工程—机械设计及理论]