小波神经网络遗传算法用于2-(9-咔唑)乙基氯甲酸酯衍生化氨基酸的胶束电动力学色谱分离优化  

Wavelet Neural Network and Genetic Algorithm for Separation Optimization of Amino Acids Derivatized with 2-(9-Carbazole) Ethyl Chloroformate by Micellar Electrokinetic Chromatography

作　　者：熊建辉[1] 张普敦[1] 郑育芳[1] 许国旺[1] 张玉奎[1] 

机构地区：[1]中国科学院大连化学物理研究所国家色谱研究分析中心,辽宁大连116011

出　　处：《分析科学学报》2004年第3期237-240,共4页Journal of Analytical Science

基　　金：国家自然科学基金(No.29975029);科技部分析测试新技术新方法研究项目(No.2002 3)

摘　　要：将小波神经网络和遗传算法应用到2 (9 咔唑) 乙基氯甲酸酯衍生化氨基酸的胶束电动力学色谱分离优化。小波神经网络结合正交试验设计用于分离过程的多因素模型建立。以训练好的小波神经网络模型为目标函数,采用实数编码的遗传算法搜寻确定最佳分离条件,在此条件下分离得到的归一化分离度积与正交试验设计中最佳条件相比,提高了12.5%。A method based on the coupling of wavelet neural network and genetic algorithm was proposed for the separation optimization of amino acids derivatized with 2-(9-carbazole) ethyl chloroformate by micellar electrokinetic chromatography. Wavelet neural network combined with orthogonal design was used to model the multi-factor affecting separation process. Using the modeled wavelet neural network as the target function, the genetic algorithm using real-valued encodings was used to search the optimal separation condition. Under the searched condition, the obtained value of normalized resolution product was increased by 12.5% compared to that in the optimal condition of the orthogonal experiment.

关 键 词：小波神经网络 遗传算法 毛细管电泳 胶束电动力学色谱 优化 

分 类 号：O657.8[理学—分析化学] Q517[理学—化学]