Research On Settlement Prediction Method Of Shallow Foundation Building Near Foundation Pit Based On I-GWO-BP

The study of surface deformation caused by excavation of foundation pits is a major research hotspot in the field of geotechnical engineering.With the increasingly complex construction environment of foundation pits and the increasing difficulty of construction,surrounding buildings(structures)are prone to greater additional settlement risks.Moreover,shallow foundation buildings are mostly brick concrete strip foundation structures with lower strength,which are more susceptible to uneven settlement leading to building tilting and damage,leading to a series of safety hazards.Therefore,based on the research on foundation pit excavation,it is particularly important to propose a method for predicting the settlement of adjacent shallow foundation buildings caused by foundation pit excavation.Through settlement prediction,it is possible to preliminarily determine whether it is necessary to protect the shallow foundation building in advance based on the predicted values before the construction of the foundation pit.This article uses neural networks to predict the settlement value of shallow foundation buildings near foundation pits.Based on finite element software combined with practical engineering,foundation pit excavation simulation is carried out,and the data items used for prediction are simulated,analyzed,and optimized to verify the feasibility of the prediction method proposed in this article in practical engineering.The main research content and achievements are as follows:(1)On the basis of previous studies on the influence of deformation caused by foundation pit excavation,the settlement of shallow foundation building near the foundation pit is taken as the research object,and 7 influencing factors causing the settlement of shallow foundation building near the foundation pit are analyzed with grey correlation degree.The results show that: The insertion ratio of supporting structure,the depth of foundation pit excavation,the number of adjacent buildings and soil weight γ have a great influence on the settlement of adjacent shallow foundation buildings.(2)Based on the BP neural network model,the settlement of the shallow foundation building near the foundation pit is predicted by taking the insertion ratio of the supporting structure,the excavation depth of the foundation pit,the number of layers of the adjacent building and the soil weight γ as the influencing factors.In order to make up for its slow convergence speed and prone to overfitting,this paper proposes to optimize BP neural network by using I-GWO(improved Gray Wolf algorithm),and compares the predicted value of BP before and after optimization with the measured value.The results show that: The I-GWO-BP model has faster convergence speed,higher accuracy,the best fitting effect,and has better applicability in predicting the settlement of shallow foundation buildings near the foundation pit.(3)Based on a foundation pit project in Zhengzhou City,ABAQUS two-dimensional numerical simulation analysis was established,and the simulated value was compared with the predicted value and measured value of I-GWO-BP.The results showed that the error between the simulated value and measured value of the shallow foundation pit building near the foundation pit was 3.2mm,and the error between the predicted value and measured value was relatively large,but the predicted value was in good agreement with the simulated maximum settlement value.(4)Based on the finite element model,the single factor sensitivity analysis is carried out on the three main influencing factors,namely the insertion ratio of supporting structure,the depth of foundation pit excavation and the number of floors of adjacent buildings.The results show that: The excavation depth of the foundation pit and the number of layers of the adjacent building have a great influence on the settlement.The insertion ratio of the supporting structure plays a greater role in the settlement of the adjacent shallow foundation pit only when the value of the insertion ratio of the supporting structure is appropriate,which has a great uncertainty.Therefore,the insertion ratio of the soil internal friction Angle φ and the supporting structure is replaced,and the I-GWO-BP prediction is carried out again.The predicted value is more consistent with the measured value,so it is feasible to predict the settlement of adjacent shallow foundation building by this method.