论文部分内容阅读
根据四川井研县坡耕地实验点对不同坡度、不同耕作方式情况下水土流失的实测资料 ,引入神经网络建模方法 ,建立了不同坡度下顺坡垄作和横坡垄作的次降雨产沙网络模型 ,得出前次降雨因子即前次降雨量、前次降雨强度及此两次降雨间隔天数对本次降雨产沙影响很大 ,不可忽略。针对复杂的非线性坡面产沙系统 ,BP网络模型比传统的回归模型能更好地刻画其本质特征。这不仅揭示了坡耕地水土流失的内在机理 ,同时为定量研究顺坡垄作改为横坡垄作的水土保持效益提供了科学依据。
According to the measured data of soil erosion at different slopes and different tillage methods, a neural network modeling method based on the experimental data of slope farmland at Jingyan County in Sichuan Province was used to establish a secondary rainfall-sediment-yielding network model , The last rainfall factor, that is, the previous rainfall, the previous rainfall intensity and the number of days between the two rainfalls have a great impact on the rainfall and sediment yield and can not be ignored. For complex non-linear slope sediment production system, BP network model can depict its essential characteristics better than the traditional regression model. This not only reveals the inherent mechanism of soil and water loss in sloping fields, but also provides a scientific basis for quantitatively studying the benefit of water and soil conservation.