反导自记忆模型的隧道沉降分析  

Analysis of tunnel settlement based on reverse self-memory model

作　　者：杨帆[1] 吕磊 田振凯 何文义[3] 

机构地区：[1]辽宁工程技术大学,辽宁阜新123000 [2]黑龙江测绘地理信息局,哈尔滨150000 [3]阜新市国土资源局,辽宁阜新123000

出　　处：《测绘科学》2017年第12期98-103,117,共7页Science of Surveying and Mapping

基　　金：国家自然科学基金项目(50604009);辽宁省"百千万人才工程"人选资助项目(2010921099);辽宁省教育厅重点实验室基础研究项目(LJZS001)

摘　　要：针对传统的变形预测模型不能对隧道高度非线性监测数据的沉降趋势和波动特征进行准确的预测问题,该文提出了反导自记忆模型。该模型运用了自记忆原理,克服了传统的变形预测模型对初值比较敏感、预测精度低等局限性,提高了对波动性数据的预测能力,之后通过工程实例验证了反导自记忆模型的可行性。最后与灰色自记忆模型进行对比,得出反导自记忆模型能够对非线性和波动性监测数据做出更加准确的预测,提高了预测的精度。In view of the problem that traditional forecasting model cannot predict tunnel subsidence trend and fluctuation characteristics of nonlinear deformation monitoring data accurately,a forecast model named as reverse memory model was presented in this paper.The model utilized the characteristic of selfmemory model,overcame the limitation of the traditional forecasting model such as be sensitive to initial value,low accuracy of prediction,and improved the ability to predict volatility data.The engineering example validations indicated that the forecast precision of the combined model was feasible.Finally,the grey self-memory model was compared with this model,and it was concluded that the reverse self-memory model could predict the nonlinear and fluctuating monitoring data more accurately,and the accuracy of the prediction was improved.

关 键 词：隧道沉降预测 自忆性原理 反导自记忆模型 沉降趋势 

分 类 号：U456.3[建筑科学—桥梁与隧道工程]