支持向量机在地下水位预报中的应用研究

针对地下水系统结构不甚清晰、基础资料不完备条件下区域地下水位预报问题,在介 绍支持向量机基本原理和实现算法的基础上,探讨了支持向量机方法在区域地下水位预报中的应用,经过与人工神经网络方法预报结果比较,表明该方法具有速度快、泛化能力强的特点,可很好地克服神经网络的过学习问题。

Application of support vector machine method in prediction of groundwater level

Aiming at the problem of predicting groundwater level under the circumstances of indistinct system structure and poor basic information, the authors introduced the support vector machine method (SVM). The basic theory and algorithm of the method were presented and application of the method to predict groundwater level was conducted. Comparison was made between SVM methods with Artificial Neural Network (ANN) and the comparative result indicated that SVM method was faster in computation and had a better generalization ability. SVM method can overcome the over-fitting problem of ANN method.