Experimental study on slurry-induced fracturing during shield tunneling

Facial support in slurry shield tunneling is provided by slurry pressure to balance the external earth and water pressure.Hydraulic fracturing may occur and cause a significant decrease in the support pressure if the slurry pressure exceeds the threshold of the soil or rock material,resulting in a serious face collapse accident.Preventing the occurrence of hydraulic fracturing in a slurry shield requires investigating the effects of related influencing factors on the hydraulic fracturing pressure and fracture pattern.In this study,a hydraulic fracturing apparatus was developed to test the slurry-induced fracturing of cohesive soil.The effects of different sample parameters and loading conditions,including types of holes,unconfined compressive strength,slurry viscosity,and axial and circumferential loads,on the fracturing pressure and fracture dip were examined.The results indicate that the fracture dip is mainly affected by the deviator stress.The fracturing pressure increases linearly with the increase in the circumferential pressure,but it is almost independent of the axial pressure.The unconfined compressive strength of soil can reflect its ability to resist fracturing failure.The fracturing pressure increases with an increase in the unconfined compressive strength as well as the slurry viscosity.Based on the test results,an empirical approach was proposed to estimate the fracturing pressure of the soil.

Assessing a soft twin tunneling numerical model using field data

Using a five-floor building affected by the Yangtze River highway tunnels in Wuhan as the engineering background, we have constructed a free-field model and a coupled model to study the soil, lining, foundations and upper structure, and analyze the rules of movements of building foundation and ground induced by single tunnel and twin tunnel excavation with the Finite Element Analysis method. It is shown that for the coupled model, the longitudinal displacement of each foundation increases slowly when the tunnel face gets close to the foundation section and then increases fast when the tunnel face moves away from the foundation during the single and twin tunneling. For a single tunnel, the surface settlements are overestimated by the free-field and coupled tunnel. This might be crucial in urban areas. Regarding the maximum settlements and the width of the settlement trough, the difference between the free-field model and the coupled model is quite obvious. This comparison with the field measurement value reveals that the coupled model seems to be superior to the free-field model. These results are of instructive significance for design and excavation.

Towards autonomous and optimal excavation of shield machine:a deep reinforcement learning-based approach

Autonomous excavation operation is a major trend in the development of a new generation of intelligent tunnel boring machines(TBMs).However,existing technologies are limited to supervised machine learning and static optimization,which cannot outperform human operation and deal with ever changing geological conditions and the long-term performance measure.The aim of this study is to resolve the problem of dynamic optimization of the shield excavation performance,as well as to achieve autonomous optimal excavation.In this study,a novel autonomous optimal excavation approach that integrates deep reinforcement learning and optimal control is proposed for shield machines.Based on a first-principles analysis of the machine-ground interaction dynamics of the excavation process,a deep neural network model is developed using construction field data consisting of 1.1 million samples.The multi-system coupling mechanism is revealed by establishing an overall system model.Based on the overall system analysis,the autonomous optimal excavation problem is decomposed into a multi-objective dynamic optimization problem and an optimal control problem.Subsequently,a dimensionless multi-objective comprehensive excavation performance measure is proposed.A deep reinforcement learning method is used to solve for the optimal action sequence trajectory,and optimal closed-loop feedback controllers are designed to achieve accurate execution.The performance of the proposed approach is compared to that of human operation by using the construction field data.The simulation results show that the proposed approach not only has the potential to replace human operation but also can significantly improve the comprehensive excavation performance.

Effects of slurry viscosity and particle additive size on filter cake formation in highly permeable sand

Filter cake is critical to maintaining the stability of the excavation face of an underwater shield tunnel in a high-permeability stratum.In a high-permeability formation,generating an effective filter cake on the excavation face is difficult with a pure bentonite slurry,which penetrates the ground and may not achieve the required suspension pressure.Determining how to efficiently and quickly form a thin and low-permeability filter cake on the tunnel working face has become a key engineering problem in the construction of slurry shield tunnels in high-permeability strata.In this study,the relationship between slurry viscosity and the slurry pressure gradient of pure bentonite was established by performing slurry permeability experiments.The influence of slurry viscosity on the formation of the filter cake in a high-permeability formation was studied under different pressure gradients.In addition,the effect of additive particle size on the slurry filter cake formation was analyzed by introducing additives with different particle sizes to pure bentonite slurries with different viscosities.The test results indicate that(1)for the pure bentonite slurry,the critical initial pressure gradient can be used as a rough indicator of the formation of the filter cake,and the relationship between the critical maximum pore diameter and the average pore diameter of the for-mation can be compared to establish and analyze the formation law of the slurry filter cake;(2)adding particles to the slurry can enhance the effect of the pure bentonite slurry;and(3)adding coarse-grained materials can effectively improve the film-forming effect.The slurry is more compact when the particle size is close to the average pore size of the formation.