关键词: 层状地基, BP神经网络, 遗传算法, 变形, 有效附加应力

Abstract: BP neural networks and genetic algorithm were combined together to establish the self-adaptive genetic algorithm and BP neural network system, and were used to predict the parameters of layered soil. Lots of physical and mechanical parameters of different layered soils are sorted out and used as the sample, then, the target parameters of layered soil were predicted. The results predicted with two kinds of intellectual technologies were compared with those predicted with BP neural networks. It shows that the ideal prediction results can be obtained simultaneously by the two technologies while variance of the sample data is small. The established system can also provide itself the generalization function to prevent the case "overfull training". When sample scale and variance of sample data are both big enough, the superiority of the network system can be better expressed.