混凝土坝位移FEM-HST混合预测模型研究

开展大坝安全监控模型的研究可以对监测异常值进行预警，有助于监测大坝的运行状态，保障其安全运行。本文基于混凝土坝的监测数据，结合有限元（FEM）分析和传统的混凝土坝位移统计分析（HST），建立混凝土坝长期运行位移FEM-HST混合预测模型，其中水压分量通过有限元计算获得，有限元模型的材料参数根据实际测点监测数据反演得到；温度分量采用简谐温度模型；时效分量采用时间的线性函数和对数函数的线性组合统计模型。所建立的FEM-HST混合模型显著改进了HST模型的预测精度和稳定性，可反映大坝和地基的结构和材料特性，能够更好预测极端荷载工况下大坝的位移，这有助于监控大坝在极端工况下的工作性态。

Study on FEM-HST hybrid prediction model for displacement of concrete dams

A dam safety monitoring model can provide an early warning of the monitored outliers, which helps monitor the operation status of a dam to ensure its safe operation. Based on the prototype monitoring data of a concrete dam and combined with analysis of finite element method (FEM) and traditional statistical analysis (HST), a FEM-HST hybrid prediction model of concrete dam displacement in long-term operation is developed in this paper. It calculates hydrostatic pressure components using a finite element model, and back-calculates material parameters from the monitoring data. Additionally, temperature components are calculated by a simple harmonic temperature model; time components by a linear combination model of a linear function and a logarithmic function. This hybrid model is a significant improvement on the HST model in prediction accuracy and model stability, better describing the structural and material characteristics of a dam and its foundation and better predicting its displacement under extreme loads. This helps monitor the working behavior of a dam under extreme conditions.