用分子力学计算出“净”原子表面积 ,并利用量子化学方法计算出化合物的电荷加权部分表面积 (CP-SA) .在用遗传算法和神经网络法对改进的 CPSA与有机醇类化合物的疏水性参数作相关分析时发现 ,改进的 CPSA可有效地用于构效关系研究 ,且算法简洁易行 ,两种多元统计方法均得到了满意的结果 . The “net” atomic surface area was calculated using molecular mechanics and the charge-weighted surface area (CP-SA) of the compounds was calculated using quantum chemistry methods. The hydrophobicity of the modified CPSA with organic alcoholates was optimized using genetic algorithms and neural networks When the parameters are analyzed, it is found that the improved CPSA can be effectively used in the study of the structure-activity relationship, and the algorithm is simple and easy. The two multivariate statistical methods are satisfied with the results.