（1 同济大学地下系 上海 200092；2 青岛理工大学 266033）
【提要】 岩土结构的位移都具有非线性动力学演化特征，针对目前采用的神经网络预测方法中存在的问题，用神经网络遗传算法耦合预测模型取代了传统的分析方法。详细介绍了建模过程，并用C语言加以实现，最后用实例论证了该方法用于基坑工程变形预测的可靠性和实用性。
【关键词】 深基坑工程 神经网络 遗传算法 变形预测
Application of Genetic Algorithm Neural Network in Deep Foundation Pit Deformation Prediction/Tao Jin1, He Keqiang2(1 Tongji University, Shanghai 200092，China; 2  Qingdao Technological University, Qingdao 266033,China)
Abstract:The method for predicting the deformation of deep foundation pit on the basis of genetic algorithm neural networks is presented. The modeling of this method is described in detail. An application software using Matlab language is developed. Finally, a practical example is given and it is proved that the predicting model is reliable and practicable.
Keywords:deep foundation pit; neural networks; genetic algorithm; deformation prediction