论文部分内容阅读
针对重力坝单测点位移监控模型存在的不足,从力学特性出发构建了重力坝挠度的多测点空间模型,并依托官地重力坝典型坝段原观监测资料分析了模型适用性及预测时间长短对模型精度影响等关键问题。结果表明,该模型较传统统计回归模型预测精度高,能通过有限测点掌控大坝变形的整体分布规律,可运用于大坝时空变形特性预测、缺测数据补齐等。同时,为保证模型预测误差控制在10%以内,建议其预测时长不超过90d。
Aiming at the shortcomings of the displacement monitoring model of single measurement point of gravity dam, a multi-measuring point space model of gravity dam deflection is constructed based on the mechanical characteristics. Based on the monitoring data of typical dam gravity dam, the applicability of the model and the prediction time length The impact of the accuracy of the model and other key issues. The results show that the model is more accurate than the traditional statistical regression model and can control the overall distribution of the dam deformation through limited measuring points. It can be applied to the prediction of the space-time deformation of the dam and the lack of padding data. At the same time, in order to ensure that the model predictive error is controlled within 10%, it is suggested that the forecasting time does not exceed 90d.