| 基于人工神经网络的红壤坡面水土流失预测模型 |
| |
| 引用本文： | 郑海金,汤崇军,汪邦稳,宋月君.基于人工神经网络的红壤坡面水土流失预测模型[J].亚热带水土保持,2009,21(4):10-13,33. |
| |
| 作者姓名： | 郑海金 汤崇军 汪邦稳 宋月君 |
| |
| 作者单位： | 1. 江西省水土保持科学研究所,江西,南昌,330029;北京林业大学水土保持学院,北京,100083
2. 江西省水土保持科学研究所,江西,南昌,330029 |
| |
| 基金项目： | 江西省科技支撑项目"红壤坡地生态农业减灾技术研究" |
| |
| 摘 要： | 基于坡面土壤侵蚀产流产沙的复杂非线性特点，引用三层前馈型BP网络建模方法，对红壤坡面土壤侵蚀产流产沙进行模拟，其中产流预测模型输入层变量数为5个，分别代表降雨量、最大30min雨强、前期降雨量、降雨动能、水保措施作用系数，输出层变量为次降雨径流损失量；产沙预测模型输入层变量数为7个，分别代表降雨量、最大30min雨强、前期降雨量、降雨动能、降雨侵蚀力、水保措施作用系数、径流量，输出层变量为次降雨土壤流失量，并利用5年田间试验所得到的坡面产流产沙实测数据，对神经网络进行模拟训练和预测，取得了较好的效果。该模型的建立与求解，为复杂的红壤坡面土壤侵蚀规律研究提供了一条新途径。
|
| 关 键 词： | 红壤坡面 BP神经网络 水土流失 |
Predicting Model of Soil and Water Loss Based on Manpower Neural Network on Red-soil Slope |
| |
| Zheng Haijin,Tang Chongjun,Wang Bangwen,Song Yuejun.Predicting Model of Soil and Water Loss Based on Manpower Neural Network on Red-soil Slope[J].Subtropical Soll and Water Conservation,2009,21(4):10-13,33. |
| |
| Authors: | Zheng Haijin Tang Chongjun Wang Bangwen Song Yuejun |
| |
| Institution: | Zheng Haijin1,2,Tang Chongjun1,Wang Bangwen1,Song Yuejun1 (1.Jiangxi Institute of Soil and Water Conservation,Nanchang of Jiangxi Province 330029 2.Soil and Water Conservation College,Beijing Forestry University,Beijing 100083) |
| |
| Abstract: | Based on the complex nonlinear characteristics of soil and water loss,the simulation was conducted to the runoff and sediment yield by the soil erosion on the slope of red soil based on the introduction of three-layer feed-forward back-propagation network model.Of which,there are five variable numbers at the input layer of runoff yield prediction model,representing five coefficients that is the rainfall,maximum 30 minutes rainfall intensity,prophase rainfall,rainfall energy and soil conservation measures fu... |
| |
| Keywords: | Red-soil slope Back-propagation Network Model Soil and Water Loss |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
