摘要
支持向量机在解决非线性及高维模式识别问题中表现出其特有的优势。针对大坝变形具有强非线性的特点以及传统神经网络模型预测时存在局部极小与过学习等问题，将支持向量机应用于大坝变形预测。以某大坝的变形监测数据为例，建立SVM预测模型，将支持向量机模型与神经网络模型进行比较分析。结果表明，该模型具有较高的精度，可以很好地应用于大坝变形监测分析。
Abstract
Support vector machine showed its specific advantage on nonlinear and high dimensional model recognition problem solving. Aiming at dam deformation had characteristic of strong nonlinear and problems like local minimum and over fitting exist in traditional neural network model prediction, support vector machine was used in dam deformation prediction. Took the deformation monitoring data of a dam as the example, SVM prediction model established, support vector machine model and neural network model were compared and analyzed. The result indicated, the precision of this model was high, which could be used in dam deformation monitoring analysis very well.
NETL
NSTL
万方数据