改进灰色理论模型在隧道施工控制中的应用  

Application of Improved Grey Theory Model in Tunnel Construction Control

作　　者：陈卓 张宏光[1] 何俊 晏长根[1] CHEN Zhuo;ZHANG Hongguang;HE Jun;YAN Changgen(School of Highway,Chang’an University,Xi’an,Shaanxi 710061,China)

机构地区：[1]长安大学公路学院,陕西西安710061

出　　处：《公路工程》2022年第6期95-101,129,共8页Highway Engineering

基　　金：国家自然科学基金项目(42077265)。

摘　　要：准确预测隧道变形发展趋势和围岩变形规律是保证施工安全的重要措施。由于深埋隧道围岩变形受到工程地质和施工等复杂因素的综合影响,具有位移序列增长的波动性和随机性。针对此特点,以灰色预测理论为基础,采用柯特斯积分公式对传统GM(1,1)预测模型的背景值求解方法进行优化,并依据相对误差平方和最小准则建立了隧道围岩变形预测的灰色模型[NCBC-GM(1,1)]。以隧道围岩变形实测值作为原始数据,用NCBC-GM(1,1)灰色预测模型对隧道围岩变形趋势进行预测,并与采用相同数据样本建模的传统GM(1,1)预测模型和基于背景值构造优化的GM(1,1)预测模型的估计和预测结果进行了对比和综合分析。结果表明,NCBC-GM(1,1)灰色模型的预测计算结果与围岩变形实测值的吻合程度相对较高,比传统GM(1,1)预测计算模型和基于背景值的构造进行优化的GM(1,1)预测计算模型均方根误差分别降低2.66、0.10 mm,平均相对误差分别降低6.76%、0.23%,预测精度高且稳定性强,能为隧道变形趋势的预测和施工控制提供重要依据。Accurate prediction of tunnel deformation development trend and deformation law of surrounding rock is an important measure to guarantee construction safety.The surrounding rock deformation of deep buried tunnel is affected by complex factors such as engineering geology and construction,so the growth of tunnel surrounding rock displacement sequence has the characteristics of fluctuation and randomness.In view of this feature,based on the grey prediction theory,the background value solving method of the traditional GM(1,1)prediction model is optimized by using the cotes integral formula.And the grey model of tunnel surrounding rock deformation prediction[NCBC-GM(1,1)]is established according to the criterion for the least square sum of relative errors.Based on the measured values of tunnel surrounding rock deformation as the original data,the NCBC-GM(1,1)grey prediction model is used to predict the deformation trend of tunnel surrounding rock.The same data samples are used for modeling,and the prediction results are compared with the traditional GM(1,1)prediction model and the GM(1,1)prediction model based on the background value construction optimization.Results show that the prediction results of NCBC-GM(1,1)grey model are in good agreement with the measured values of surrounding rock deformation.Compared with the traditional GM(1,1)prediction model and the GM(1,1)prediction model based on the background value construction optimization,the root mean square errors of the two models are reduced by 2.66 mm and 0.10 mm,and the average relative errors are reduced by 6.76%and 0.23%,respectively.The NCBC-GM(1,1)grey model has high prediction accuracy and good stability,which can provide important basis for tunnel deformation trend prediction and construction control.

关 键 词：隧道工程 变形预测 灰色理论 背景值 边界修正 

分 类 号：U455[建筑科学—桥梁与隧道工程]