| In the deep foundation pit project in the middle and lower reaches of Yangtze River plain,due to the large amount of rainfall,the water table is usually high,and the soil is mostly water-rich soft clay,the soil is affected by rainfall and groundwater seepage,which often has a huge impact on the deformation of the enclosure structure during the excavation and support of the foundation pit.Research shows that the seepage field formed by rainfall and groundwater and the stress field formed by soil interact with each other,and under the action of complex seepage field,the change of pore water pressure and infiltration path often leads to the change of effective stress of soil,therefore,the pit monitoring of single study of stress field and displacement field is not enough to characterize the change law of coupled field,and,at present,the pit monitoring mostly adopts manual point type field monitoring,with scattered monitoring points.The data analysis often adopts manual post-processing,which has a long analysis period and large errors,and cannot realize the real-time effective monitoring of the negative environmental effects and the bearing characteristics of the enclosure structure during the excavation of the foundation pit.In view of this,the coupled field theory analysis of the whole foundation pit and real-time monitoring based on artificial intelligence technology become the urgent problem of foundation pit excavation safety.This paper takes the excavation and support system of foundation pit in coupled field as the research object,and integrates theoretical analysis,physical test and numerical simulation to conduct a systematic study on the soil deformation of foundation pit under the action of coupled field and its implementation of monitoring and early warning technology.Firstly,based on theoretical analysis,a fluid-solid coupling model is established,and the basic physical and mechanical properties of the pit soil are obtained through indoor tests;secondly,based on the index parameters under different conditions obtained from indoor tests,the stability of the excavation and support stage of the pit is analyzed by relying on Plaxis finite element software,and the deformation law of the excavation and support of the pit under the action of coupled fields and the threshold value under the most unfavorable conditions are obtained;finally Finally,artificial intelligence is adopted to apply research on the stability of foundation pit excavation,based on wireless sensors to automatically collect site monitoring data,compare and analyze the collected data with the soil deformation law in the coupled field obtained by finite element simulation,and establish a safety early warning system for foundation pit excavation support based on artificial intelligence technology.The main research contents are as follows:(1)Based on the seepage and consolidation theories,the interaction between seepage and stress fields was analyzed,a mathematical model of the coupling of seepage and stress fields was established,the steps for solving the mathematical model were determined,and the accuracy of the mathematical model was verified.(2)Through the indoor geotechnical tests,the basic physical and mechanical parameters of the soil,including water content,density,liquid-plastic limit,maximum dry density,optimum water content,cohesion and internal friction angle,were obtained,which provided the basic conditions for the establishment of the finite element model.(3)The deformation law of the foundation pit excavation support system in the coupled field was obtained based on finite element simulation.It was found that: when the water head increased from-8m to-2m,the overall deformation of the foundation pit increased by 35.5%;when the permeability coefficient increased from 0.005m/day to 0.08m/day,the overall deformation of the foundation pit increased by 17.62%;when the foundation pit increased from single-layer soil to multi-layer soil,the overall deformation of the foundation pit increased by 11.4%.Various factors The effect of various factors on the deformation of the pit is as follows: water table height > permeability coefficient > excavated soil layer.High water level,large permeability coefficient and multi-layered soil body are the most unfavorable factors for the excavation support of the foundation pit.At the same time,the critical value of pit damage can be obtained when several unfavorable factors obtained based on finite element simulation exist at the same time,and it is used as the early warning threshold for the deformation of pit excavation support system under the action of coupled field.(4)Based on finite element simulation,the coupled field and uncoupled field are compared and studied,and the optimization analysis of the pit excavation and precipitation scheme in the coupled field is carried out.The results show that,from the overall deformation point of view,whether it is the pit of the pile internal support system or the pit of the anchor rod external support system,the pit deformation under the action of the coupled field is larger than that under the action of the uncoupled field,and the maximum deformation increment is59.7% and the minimum deformation increment is 11.7%.At the same time,the deformation of the pit under the action of coupled field is reduced by adjusting the depth of precipitation,so that it is effectively controlled within the warning threshold,and then the optimal plan of pit excavation and precipitation is obtained.It is found that when the change of precipitation depth is 0.5m,the deformation in the pit is the smallest,which provides the optimal precipitation scheme for the subsequent actual construction.(5)Based on the Damao interchange project,an intelligent and information-based foundation pit monitoring and early warning cloud platform is developed,and automatic monitoring means are used to monitor the excavation construction in real time.Based on the deformation law and early warning threshold of foundation pit excavation support in the coupled field obtained by finite element simulation,and the measured data are automatically compared and analyzed with the early warning threshold obtained by finite element simulation,so as to achieve the effect of real-time early warning. |