论文部分内容阅读
文章简要的分析了影响边坡冲刷量的因素,结合神经网络模型对边坡土冲刷量计算进行了模型模拟,并通过试验段的证明说明此模型的模拟效果较好。结果表明:影响冲刷量的因素是复杂多样的,且这些因子之间相互作用,既存在输入输出为前向传播的关系,又存在某种内在的反馈关系,这种关系正符合神经网络模拟;由于以往利用观测资料建立经验回归公式难以描述各冲刷因子间复杂的关系,所以利用神经网络计算具有较好的概括性和实用性。
In this paper, the factors influencing the amount of slope erosion are briefly analyzed, and the calculation of the amount of erosion on slope soil is carried out by combining with the neural network model. The result of the test indicates that the model has good simulation results. The results show that the factors that affect the erosion amount are complex and diverse, and the interaction between these factors exists in both the input and output for the forward propagation and the existence of some intrinsic feedback relationship, which is in line with the neural network simulation; Due to the fact that it is difficult to describe the complex relationship among the scouring factors by using empirical data to establish empirical regression formula, the neural network has better generalization and practicability.