摘要
介绍了人工神经网络 BP模型和 MATLAB神经网络工具箱。在分析坡面冲刷过程与神经网络模型特点具有某些相似的基础上 ,以 MATLAB- NNT为工具对 6种模型结构进行试验 ,证明 ANN4效果最好。利用 ANN4对土的冲刷量进行模拟及预测 ,通过与连徐高速公路邳州段路堤边坡冲刷试验观测资料分析比较 ,显示具有较好的模拟预测效果。
The artificial neural network, the back propagation model and the MATLAB-NNT are described. Based on the similarity between the process of highway slope erosion and ANN, and on the MATLAB-NNT, six models are tested. The result shows that effect of ANN4 is the best. The soil erosion is simulated and forecasted by ANN4. The analysis comparison shows an ideal effect between the simulated soil erosion and the observation data from the road bank slope erosion experiment in Pizhou segment of Lian-Xu highway.
出处
《解放军理工大学学报（自然科学版）》
EI
2003年第5期75-77,共3页
Journal of PLA University of Science and Technology(Natural Science Edition)
