欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
支持向量机拟合模型（SVR）可用于大坝裂缝预报,但自变量间的多重相关性和输入变量的高维数对支持向量机拟合模型的精度影响较大。基于大坝裂缝开合度理论,利用主成份分析法（PCA）提取原样本信息缩减后的主成分作为SVR模型的输入量,构建了大坝裂缝开合度的早期预报PCA-SVR模型。将该模型应用于某大坝监控资料的分析中,与传统回归模型相比,PCA-SVR模型具有更高的计算精度和运算效率,并可提前预报裂缝开合度信息,能在实际工程中广泛应用。
英文摘要：
The SVR model can be used in dam crack prediction,but the multiple correlations among independent variables and high dimension of input variables have large impact on the accuracy of support vector machine fitting model. On the basis of dam crack opening theory, the Principal Component Analysis method is adopted to extract the information of original sample, and the obtained main components are used as the input variable of SVR model, which can reduce the computational cost. The PCA - SVR model was established. This new model is applied to the analysis of monitoring data, and the results show that the PCA - SVR model has higher calculation accuracy and efficiency, and can predict the dam opening information compared with traditional regression methods.
同期刊论文项目
同项目期刊论文
期刊信息
位置：