论文部分内容阅读
分析了堤坝管涌发生的机理和影响堤坝管涌发生的重要因素,选取了8个实测指标作为预测依据,建立了距离判别分析模型和支持向量机法的堤坝管涌预测方法,并和神经网络方法进行了对比分析。通过对23个堤防管涌工程实例的研究表明,距离判别法和支持向量机法预测模型性能良好。基于神经网络核函数和径向基核函数的支持向量机具有更高的预测精度,支持向量机法是解决堤坝管涌预测问题的有效方法,可以在实际中应用。
Based on the analysis of the mechanism of dykes and the important factors that affect the occurrence of dikes, eight measured indexes are selected as the basis of prediction, and the dyke piping prediction method based on distance discriminant analysis model and support vector machine method is established. And neural network method Comparative analysis. The study of 23 embankment piping projects shows that the distance discriminant method and the support vector machine method have good performance. Support vector machines based on neural network kernel function and radial basis kernel function have higher prediction accuracy. Support vector machine (SVM) is an effective method to solve the problem of embankment collapse prediction and can be applied in practice.