飞来峡大坝变形性态分析与预报模型研究

通过对飞来峡大坝的水平位移进行分析，建立相应的统计模型，并通过统计模型与BP神经网络模型进行比较，建立相应的预报模型，为飞来峡大坝的安全监测提供依据．通过分析比较得到结论：飞来峡大坝目前变形性态正常，水平位移年变幅主要受温度荷载的影响，BP神经网络模型虽然拟合精度较统计模型高，但是其预测能力并不优于统计模型，而且统计模型能对监测数据作分解和成因解释，所以采用统计模型对飞来峡大坝建立了预测模型。

Deformation analysis of Feilaixia Dam and study of forecast model

The horizontal deformation of Feilaixia Dam was analyzedin this paper. Then the statistical model was developed, and through a comparison between statistical model and BP neural network model, the corresponding forecast model was given, which is the safety monitoring guide of Feilaixia Dam. According to above analysis, some results obtained: the deformation of Feilaixia Dam is normal. The thermal load is the main load to the deformation range of horizontal deformation. Though the regress precisionof BP neural network model was higher than the statistical model, the forecast precision of the former was not better than the latter, and the latter is suitable for analyzing and explaining the obtained monitoring data, the forecast model of the Feilaixia Dam was developed by statistical model.