建立了基于免疫遗传算法的人工神经网络结构,用于预测α-和β-环糊精与单取代或1,4-二取代苯衍生物包结物的稳定常数,其中神经网络用于建立取代苯中2个取代基的摩尔折射率R_m,疏水性参数π和Hammett常数σ与包结物稳定常数之间的QSAR模型,而免疫遗传算法则用于优化网络的权重系数。结果表明,由于免疫机制的引入使遗传算法的优化效率提高,神经网络的学习功能得到明显改善。 An artificial neural network structure based on immune genetic algorithm was established to predict the stability constants of α-and β-cyclodextrin inclusion complexes with monosubstituted or 1,4-disubstituted benzene derivatives. The neural network was used to establish the substitution The molar refractive index Rm, the hydrophobic parameter π and the Hammett constant σ of the two substituents in benzene and the QSAR model between inclusion stability constants, while the immune genetic algorithm is used to optimize the weight coefficient of the network. The results show that the learning function of neural network has been significantly improved due to the improved efficiency of genetic algorithm due to the introduction of immune mechanism.