摘要
采用粒子群算法、广义回归神经网络和差分自回归移动平均模型实现基坑变形预测,评价基坑施工过程中的稳定性,以佐证基坑支护措施的有效性。基坑区地质条件、近接条件均较为复杂,需采用刚度大的支护结构来控制基坑变形,因此,将基坑支护形式确定为桩锚支护结构,且为确保基坑支护体系的经济性,还进行了分区支护措施设计。在基坑稳定性评价过程中,现阶段条件下的沉降变形的变化范围为12.68~21.42 mm,均值为17.84 mm,水平位移的变化范围为15.08~22.49 mm,均值为18.68 mm,得出两类变形均小于变形控制值,说明基坑目前的变形在变形控制值范围内,其稳定性较好;通过变形预测,得出基坑后续变形具收敛趋势,后续的变形值范围为21.35~23.58 mm,也小于变形控制值,说明基坑后续的变形控制也较好,基坑稳定性较优。
Particle swarm optimization algorithm,generalized regression neural network,and differential autoregressive moving average model are used to predict the deformation of foundation pits.The stability of the foundation pit construction process is evaluated based on the prediction results to demonstrate the effectiveness of foundation pit support measures.The geological conditions and proximity conditions in the target excavation area are relatively complex,requiring the use of rigid support structures to control the deformation of the foundation pit.Therefore,pile anchor support structure is chosen as the form of foundation pit support.To ensure the economic efficiency of the foundation pit support system,zoning support measures are also designed.In the process of evaluating the stability of the foundation pits,the range of settlement deformation under current conditions is 12.68-21.42 mm,with an average of 17.84 mm.The range of horizontal displacement change is 15.08-22.49 mm,with an average of 18.68 mm.It is found that both types of deformation are smaller than the deformation control value,indicating that the current deformation of the foundation pit is within the deformation control value range and has good stability.Through deformation prediction,it is found that the subsequent deformation of the foundation pit has a convergence trend,and the range of subsequent deformation values is 21.35-23.58 mm,which is also smaller than the deformation control value.This indicates that the subsequent deformation control of the foundation pit is also good,and so is the stability of the foundation pit.
作者
李锐豪
LI Ruihao(Henan No.2 Institute of Geological and Mineral Resources Survey Co.,Ltd.,Zhengzhou 450000,Henan,China)
出处
《城市地质》
2024年第2期225-232,共8页
Urban Geology
关键词
基坑
支护设计
变形监测
广义回归神经网络
变形预测
foundation pit
support design
deformation monitoring
generalized regression neural network
deformation prediction
