摘要
针对大断面碳化泥质板岩隧道在施工过程中极易发生塌方、拱架扭曲变形以及频繁换拱的特点,在小盘岭隧道进出口段选择多个监测断面作为试验段进行现场试验研究。分别对净空收敛变形、拱顶下沉、围岩压力、锚杆轴力、喷射混凝土应变、钢拱架支撑应力、二次衬砌混凝土应变、二次衬砌背部围岩压力进行测试。研究结果表明,隧道持续变形一般到48 d才能稳定,并且累计变形量很大,拱顶沉降最大能达到334 mm,累计收敛值最大能到318 mm。在控制围岩变形方面,对支护参数的监测发现:采取0.5 m的开挖步距,采取I20钢支撑,以及采取隧道衬砌背后的径向注浆,能有效的降低围岩变形及衬砌结构的受力。同时利用BP神经网络方法,能够比较准确的预测隧道围岩在施工过程中的拱顶沉降及围岩收敛,对隧道的开挖具有较大的指导作用。
Considering landslides,distortion of steel arch and frequent arch changing occur easily for large cross section tunnels of carbide argillite in the course of construction,Selected multiple monitor sections near the entrances and exits of Xiaopanling tunnels for the field test. Conducted the convergence deformation test,vault subsidence test,bolt axle force test,sprayed concrete strain test,steel camber bearing stress test,secondary lining concrete strain test,secondary lining back surrounding rocks stress test respectively. The results show that the tunnel deformation stabilizes at 48 d and the deformation accumulation is large,the vault subsidence reached 334 mm,convergence value reached 318 mm. The study also shows that taking the excavation step of 0.5 m,adopting I20 steel bracing,and taking the radial grouting behind the tunnel lining,can effectively reduce the stress of surrounding rock deformation and lining structure. This study points out the effects of the network on veracity and advantages of deformation prediction of surrounding rock. This method has great guidance for the excavation of tunnel construction site.
出处
《岩石力学与工程学报》
EI
CAS
CSCD
北大核心
2016年第A02期4029-4039,共11页
Chinese Journal of Rock Mechanics and Engineering
基金
国家重点基础研究发展计划(973)项目(2010CB732102)
北京交通大学基本科研业务费项目(2011JBM078)~~
关键词
隧道工程
软弱地层
大断面隧道
拱顶沉降
收敛
BP神经网络
tunnelling engineering
soft ground
large section tunnel
arch settlement
convergence
BP artificial neural network
