论文部分内容阅读
针对水土流失预测的多因素及非线性等特征,引入自适应模糊神经推理系统(Adaptive network-based fuzzy inference system,ANFIS),对自然降雨条件下不同生态措施的水土流失规律进行了预测。结合MATLAB模糊推理工具箱,以试验站2001—2005年的51场降雨为训练样本进行学习训练,以2008年观测的10场降雨为检验样本进行验证,结果表明:4个处理(百喜草全园覆盖、百喜草覆盖果园、清耕果园和裸露对照)的径流预测的平均相对误差的绝对值分别为17.04%、18.89%、15.74%和16.61%,平均17.07%;泥沙的预测平均误差的绝对值为20.87%、17.71%、19.28%和16.71%,平均为18.57%;从误差分析可以看出模型具有较高的精度和稳定性,表明ANFIS模型可以有效的描述自然降雨条件下不同生态措施坡地的水土流失特征,而且还可以描述和处理模糊信息。ANFIS模型不但操作简单,计算量小,而且在物理参数较少的情况下也可以取得较高的精度,为自然降雨条件下坡地水土流失预测的进一步研究提供了新的思路。
According to the multi-factor and non-linear characteristics of soil erosion prediction, adaptive network-based fuzzy inference system (ANFIS) is introduced to predict the law of soil erosion under different natural rainfall conditions. Combined with MATLAB fuzzy inference toolbox, 51 rainfalls from 2001 to 2005 in the experimental station were studied and trained. The results of 10 rainfalls observed in 2008 were tested. The results showed that 4 treatments The average relative errors of the runoff forecast of runoff forecasting were 17.04%, 18.89%, 15.74% and 16.61%, respectively, with an average of 17.07%. The average forecast error of sediment The absolute value of the model is 20.87%, 17.71%, 19.28% and 16.71%, respectively, with an average of 18.57%. From the error analysis it can be seen that the model has high accuracy and stability, indicating that the ANFIS model can effectively describe the different ecological conditions under natural rainfall Measures the characteristics of soil and water loss on the slope, but also can describe and deal with fuzzy information. The ANFIS model not only has the advantages of simple operation and small calculation, but also achieves higher precision with less physical parameters, which provides a new idea for further research on predicting soil and water loss in slope under natural rainfall.