李家峡拱坝参数反演分析与安全监控模型研究-水利水电工程专业论文.docxVIP

摘 摘 要 李家峡水电站从1996年12月下闸蓄水以来，历经从初期蓄水位到电站运 行最高水位的历程，水电站各水工建筑物枢纽已经历了正常高水位的检验，建筑 物的结构形态经过不同水位工况下的运行调整，必将发生较大的改变。及时了解 和掌握水工建筑物的工作状态，是建筑物能够长久安全稳定运行的保障。从国内 外各有关水工建筑物运行管理的经验来看，随着建筑物运行时间的增长，水工建 筑物及基础的有些力学参数将会发生一定的调整和改变。作者利用李家峡水电站 在正常高水位运行时所监测到的原型观测资料，用有限元模型计算大坝在各种不 同工况下不同监测点部位的变位值，然后建立基于人工神经网络的大坝及地基的 参数反演计算模型，再用垂线原型观测资料来反演大坝及地基的参数(坝体弹性 模量、温度线膨胀系数及坝基变形模量)，并进行相应的分析评价。另外，用已有 的原型观测资料成果分析计算大坝变形监测的混合模型及统计模型。本文的主要 工作和取得的成果如下： 1-根据实际地质情况建立李家峡大坝的有限元计算模型。 2．在广泛查阅有关文献的基础上，对大坝变位的正分析计算和人工神经网 络用于监测资料反分析的原理进行了总结，建立了用有限元计算与人工神经网络 相结合进行大坝及基岩参数反演的反分析模型。 3．用人工神经网络建立坝体变位的预测模型。 4．根据现有的原型观测资料建立坝体变位的混合模型及统计模型。 通过以上工作可以认为，用人工神经网络来建立坝体变位的神经网络模型 及对大坝的参数进行反演计算的理论和方法是可行的，计算过程快速、简便，适 用性较强，可为大坝安全监测资料分析提供另一种方法和途径。 关键词： 李家峡水电站安全监控模型参数反演观测资料分析 ABSTRACTSince ABSTRACT Since December,1996 when the sluice gates were closed and water storage began， Lijiaxia Hydropower Station has experienced the whole process from initial storage level to the maximum pool level．The hydraulic structures of the hydro-project have also experienced normal hi班stage．After operation in cases of different water levels structural forms of the hydro-structures have been adjusted and changed significantly． In order to ensure the hydraulic shuctures to safely and stably operate for a long time， the working state of the hydro-struct叶es should be understood and mastered in time． Based on the experianee of operation and management of relative hydro-structures home and abroad，along with extension of operation time of the hydro-structure，some mechanical parameters of the hydro·structure and its foundation will undergo a certain degree of adjustment and change．On the basis of data from prototype observation which were monitored during the Iaji缸ia Hydmpower Station was operated at normal high啪te￡level the author calculated displacements at different monitoring spots under different a隅b擎璐e of FEM model,then based on the aftificial neural network the back calculation model of the dam and its foundation Was established． Afterward,data from vertical