Parameter Inversion Of Soil Layer Using YEBL And Its Application In Settlement Analysis Of Tunnel-Before-Station Construction

With the rapid and high-quality development of the national economy and the gradual completion of the urban transportation network,the construction methods of various subway stations have also become a key research field."Tunnel before station" construction method is a kind of construction method for constructing subway stations,which uses shield tunneling to cut through interstation tunnels and then conducts underground excavation of subway stations.It has the characteristics of short construction period,flexible construction,and strong antiinterference;However,due to the excavation of shield machines and the large range of underground excavation construction,the strata are greatly disturbed,resulting in dynamic changes in engineering geotechnical parameters.In response to the above issues,in order to study the dynamic change process of geotechnical parameters and compensate for the problem that the geotechnical parameters obtained by traditional geological survey methods cannot fully reflect the geotechnical properties,this paper combines numerical simulation with swarm intelligent optimization algorithms to obtain practical geotechnical parameters under various conditions during the "tunnel before station" construction process through engineering geotechnical parameter inversion.The main research content and results of this article are as follows:(1)Aiming at the construction process of the expanded excavation section of the left and right line platform tunnels of a subway station project in Guangzhou,a two-dimensional finite element model using ANSYS was established.Combined with the monitoring data of surface settlement during the construction process,the impact of construction on surface settlement was analyzed,and the law of ground disturbance caused by the construction of the tunnel before the station was revealed.(2)Aiming at the shortcomings of swarm intelligence optimization algorithms,an improved empirical learning based swarm intelligence optimization algorithm(YEBL)is proposed by introducing the idea of yin and yang pairs.By comparing and analyzing different swarm intelligence algorithms through multiple test functions such as unimodal,multimodal,and CEC 2017,it is successfully verified that the improved empirical learning based swarm intelligence optimization algorithm has faster search speed and better global search ability.(3)Based on the displacement monitoring data obtained from limited key locations,combined with the two-dimensional finite element model of tunneling before stationing,the improved optimization algorithm in Chapter 3 is used for back analysis of geotechnical parameters.By comparing the inversion results of geotechnical parameters using the improved empirical learning algorithm before and after the improvement,it is verified that the YEBL algorithm has faster iterative convergence than the EBL algorithm,Applying this algorithm to the back analysis of geotechnical parameters in the "tunnel before station" construction project is more reliable and accurate.(4)Based on the inversion results of soil layer parameters and in combination with the different excavation and construction stages of the tunnel to station project,input the dynamic soil layer change parameters during construction,establish a global three-dimensional finite element model for subway station projects including platform tunnels,cross passages,etc.,reveal the rules of ground disturbance caused by the tunnel to station construction and the impact on various structures of the subway station,and analyze the stability and safety of the entire construction process.