龙羊峡重力拱坝水平位移监测资料统计模型对比分析  

Comparative Analysis of Statistical Models of Horizontal Displacement Monitoring Data of Longyangxia Gravity Arch Dam

作　　者：陈勋辉 Chen Xun-hui(Dam Management Center of State Power Investment Group Co.,Ltd.Xi'an 710000,China;Qinghai Yellow River Upstream Hydropower Development Co.,Ltd.,Xining 810000,China;College of Water Conservancy and Environment,Three Gorges University,Yichang 443002,China)

机构地区：[1]国家电力投资集团有限公司大坝管理中心,西安710000 [2]青海黄河上游水电开发有限责任公司,西宁810000 [3]三峡大学水利与环境学院,湖北宜昌443002

出　　处：《西北水电》2024年第3期89-94,共6页Northwest Hydropower

基　　金：国家自然科学基金项目(51209124)。

摘　　要：影响拱坝变形最为重要的因素为气温和水压,因此气温及其滞后效应是改进拱坝变形统计模型主要考虑因素之一。传统HST(hydraulic-seasonal-time,水压-气温-时间)模型中,气温分量常采用一系列简单的三角函数来模拟,至于实测气温及其滞后效应对统计模型精度的影响有待进一步研究。以龙羊峡大坝为例,选择垂线典型测点,首先基于实测资料建立HST模型,其次采用实测温度代替HST模型中基于三角函数所模拟的温度分量,建立HTT统计模型,最后提出了一种HSTT(水压、季节、温度和时间)混合预测模型,该模型综合考虑了HST和HTT里边所有变量(如季节函数、实测气温、气温滞后因子)。结果表明:与HTT模型相比,HSTT模型精度更高,说明三角函数对温度项模拟也有其不可替代的作用;引入实测气温数据后,统计模型精度均有所提升;HTT模型和HSTT模型中,对气温做移动平均处理比做主成分提取作为输入效果更佳。The most important factors affecting the deformation of arch dam are air temperature and water pressure.Therefore,air temperature and its lag effect are one of the main factors to improve the statistical model of arch dam deformation.In the traditional hydraulic-seasonal-time(HST)model,the temperature component is often simulated by a series of simple trigonometric functions.The influence of the measured temperature and its lag effect on the accuracy of the statistical model needs to be further studied.Taking the Longyangxia dam as an example,typical vertical measuring points are selected.First,the HST model is established based on the measured data.Secondly,the measured temperature is used to replace the temperature component simulated by trigonometric functions in the HST model to establish the HTT statistical model.Finally,a HSTT(water pressure,season,temperature and time)hybrid prediction model is proposed,which takes into account all the variables in HST and HTT(such as seasonal function,measured temperature and temperature lag factor).The results show that compared with the HTT model,the HSTT model has higher accuracy,indicating that trigonometric functions also have an irreplaceable role in simulating temperature items;after introducing the measured temperature data,the accuracy of the statistical model has been improved;in the HTT model and HSTT model,the moving average treatment of the temperature is better than the principal component extraction as input.

关 键 词：统计模型 气温因子 HST 模型 

分 类 号：TV698.1[水利工程—水利水电工程]