对于传统的大坝变形监控模型,温度分量有着欠拟合的现象,对此进行了改进。为了反映大坝整体变形规律,引入空间坐标向量,建立了基于温度分量改进的大坝变形空间函数监控模型。所建的监控模型由于因子较多,因此应用量子遗传算法及逐步回归分析选择了显著性因子并剔除了不显著因子。最后将所建的大坝变形空间函数监控模型应用于某混凝土坝变形监测中,结果显示,所建模型较传统大坝变形监控模型精度高。
The under fitting of temperature component in traditional monitoring model of dam deformation was studied in this paper and temperature component was improved. In order to reflect the integrally deformation law, spatial coordinate vector was introduced to establish the spatial function monitoring model of dam deformation considering the improvement of temperature component. However, the established monitoring model had dozens of factors. Therefore, quantum genetic algorithm and stepwise regression analysis were adopted to select significantly factors and remove non-significantly factors. At last, the spatial function monitoring model of dam deformation was applied to the deformation monitoring of a concrete dam, and the results show that the established monitoring model has better fitting precision relative than that of the traditional monitoring model.