Surface Subsidence Research Articles

Boosting Model Interpretability for Transparent ML in TBM Tunneling

Tunnel boring machines (TBMs) are essential for excavating metro tunnels, reducing disruptions to surrounding rock, and ensuring efficient progress. This study examines how machine learning (ML) models can predict key tunneling outcomes, focusing on making these predictions clearer. Specifically, the models aim to predict surface settlements (ground sinking) and the TBM’s penetration rate (PR) during the Athens Metro Line 2 extension to Hellinikon. For surface settlements, four artificial neural networks (ANNs) were developed, achieving an accuracy of over 79%, on average. For the TBM’s PR, both an XGBoost Regressor (XGBR) and ANNs performed consistently well, offering reliable predictions. This study emphasizes model transparency mostly. Using the SHapley Additive exPlanations (SHAP) library, it is possible to explain how models make decisions, highlighting key factors like geological conditions and TBM operating data. With SHAP’s Tree Explainer and Deep Explainer techniques, the study reveals which parameters matter most, making ML models less of a “black box” and more practical for real-world metro tunnel projects. By showing how decisions are made, these tools give decision-makers confidence to rely on ML in complex tunneling operations.

Research on the Influence of Shallow Buried Tunnel Crossing on the Stability of Overlying Frame Structure Building

The land section of the Huangdao end of the Jiaozhou Bay Second Submarine Tunnel is extensively underlain by the Quaternary Loose Accumulation Layer. The tunnel passes through a weathered granite fracture zone with well-developed rock joints beneath the buildings. The tunnel excavation process significantly disturbs the buildings above, making them prone to settlement, cracking, and tilting. This research conducts numerical simulations of three tunnel excavation methods, and based on the results, compares the deformation behaviors of the ground surface and buildings under various conditions. The findings show that the double side-wall guide pit method has better adaptability in controlling surface settlement and building deformation than the vertical or curved CD method. Moreover, the removal of temporary supports significantly affects building settlement and tilt; this risk can be effectively reduced by controlling the stress relief ratio during the removal phase of the temporary supports in the tunnel. The significance of the study lies in the fact that by choosing an appropriate tunnel excavation support scheme, the disturbance impact on the overlying buildings can be minimized, and the construction safety and stability of the surrounding buildings can be guaranteed. The results of this study can provide initial guidance for constructing shallow-buried tunnels beneath existing buildings.

Study on surface subsidence law in mining area based on SBAS-InSAR technology

In view of the problems of traditional mine monitoring technology, which is time-consuming, laborious, small amount of monitoring data, and affected by weather, this paper selects radar satellite earth observation technology, adopts SBAS-InSAR image processing method, obtains the surface settlement changes in the study area, and analyzes several typical settlement funnels in the mining area: the data results show that the maximum settlement in the Fengyuan Village mining area is more than -30mm, and the maximum settlement in the Loess Temple mining area and Shagou mining area is more than -50mm. The Kentielinggou mining area and the Heishiyan mining area have the fastest settlement at two points, with a maximum settlement of more than -90mm.

Research on prediction of high energy microseismic events in rock burst mines based on BP neural network

In response to the frequent occurrence of high-energy microseismic events in coal mines in China, a back propagation neural network (BPNN) prediction model based on surface subsidence data has been proposed to provide a basis for safely and efficiently predicting coal mine disasters. Theoretical research on the relationship between surface displacement, mining disturbance, and high-energy microseismic event levels has demonstrated a significant correlation among these factors. When there is a sudden increase or decrease in surface displacement or mining disturbance, the advancing working face typically exhibits dynamic characteristics. Therefore, feature parameters relevant to predicting high-energy microseismic event levels were selected as input variables for the BPNN model. Raw data from 88 sets of microseismic events in the 301 working face of the third mining area of a certain coal mine in Inner Mongolia were extracted. First, outlier preprocessing was conducted to obtain a complete dataset, which is then divided into a training set and a testing set. The BPNN model was trained with the training set and subsequently tested to evaluate its predictive performance. Finally, by comparing several model evaluation metrics, it was found that the BPNN model outperforms other common models in predicting high-energy events. The overall prediction accuracy is 86.4%, and the root mean square error (RMSE) is 0.45, indicating that the BPNN-based prediction model for high-energy microseismic events in coal mines is feasible.

Tundra fires and surface subsidence increase spectral diversity on the Yukon–Kuskokwim Delta, Alaska

Abstract Tundra fires can dramatically influence plant species cover and abundance, organic layer depth, and the magnitude of seasonal permafrost thaw. However, knowledge of the impact of wildfire on short and long-term interactions between vegetation and permafrost thaw remains limited. Here, we evaluate the spatial and temporal interactions between wildfire disturbance and surface subsidence on a remotely derived proxy for species diversity (i.e. spectral diversity (SD)) of 16 fire scars within the Izaviknek and Kingaglia uplands of southwestern Alaska’s Yukon–Kuskokwim Delta with burn dates between 1971 and 2015. SD was calculated as the sum of squared spectral variance of pixel spectra from the mean spectra, within a plant community (analogous to alpha diversity), between plant communities (beta diversity), and across terrain composed of a mosaic of communities (gamma diversity). Surface subsidence was calculated from spaceborne interferometric synthetic aperture radar data from Sentinel-1. Results indicate the burn scars had consistently lower total gamma diversity and greater rates of subsidence than paired unburned reference areas, where both gamma diversity (R 2 = 0.74, p < 0.001) and relative subsidence (R 2 = 0.86, p < 0.001) decreased with the time since burn. Compared to older burn scars, young scars had higher gamma spectral diversities (0.013 and 0.005) and greater subsidence rates (−0.097 cm day−1 and −0.053 cm day−1). Communities subsiding at higher rates had higher gamma diversities (R 2 = 0.81, p < 0.001). Results indicate that rates of post-fire vegetation succession are amplified by the thickening of active layers and surface subsidence that increases both spectral and species diversity over 15 years following fire. These results support the idea that SD may be used as a remotely sensed analog of species diversity, used to advance knowledge of the trajectories of plant community change in response to interacting arctic disturbance regimes.

Explainable deep learning-based dynamic prediction of surface settlement considering temporal characteristics during deep excavation

Accurate prediction of surface settlement (SS) induced by deep foundation pit (DFP) excavation is challenging considering the complicated soil distributions and uncertain construction situations. Therefore, a Recurrent Gated Unit (GRU) neural network model incorporated with Variation Mode Decomposition (VMD) was developed for SS prediction. By data denoising with VMD and hyperparameter tuning using Bayesian Optimization (BO), the proposed BO-GRU model achieved dynamic and accurate prediction of SS caused by DFP excavation. Shapley Additive exPlanations (SHAP) analysis was then performed to enhance the interpretability of the BO-GRU model. The presented BO-GRU model was validated with a case study of Wuhan Metro Line 12 Shiqiao Station. The results indicate that: (1) The BO-GRU models with VMD data denoising technique achieved accurate prediction with the average RMSE of 0.0244, MAE of 0.0189, MAPE of 0.0017, R2 of 0.9894, and VAF of 99.1206. (2) The BO-GRU model outperformed the other five state-of-the-art models in predictive performance with robustness improvement of 13.5%, 12.0%, 50.0%, 29.2%, and 27.3%, separately, compared to the other five models. (3) The historical values of the SS make the most significant contributions to the outcomes of the BO-GRU models. In short, this study enhances the accurate and dynamic prediction of SS caused by DFP excavation, contributing to the safe execution of DFP projects.

Machine Learning–Finite Element Mesh Optimization-Based Modeling and Prediction of Excavation-Induced Shield Tunnel Ground Settlement

Ground settlement prediction for shield construction is highly important and challenging. This study introduces a machine learning algorithm combined with finite element numerical simulation, i.e., machine learning–finite element mesh optimization. For surface subsidence prediction, 16 combination models of ANN, KNN, RF and SVR were optimized by PSO, GA, BT and BO, involving raw data preprocessing, principal component analysis, hyperparameter selection and prediction accuracy evaluation. A subway shield tunneling project was analyzed, in which the meshes of finite element numerical models were discretized into different sizes from 1.0[Formula: see text]m to 2.0[Formula: see text]m. In total, 360 sets of data points were extracted from the simulation results, including stress, strain, shield jacking force, internal friction angle, cohesion force, and settlement, of which 252 data points were used as the input parameters of machine learning model. Analysis of average error rate of finite element–machine learning coupling models showed that the finite element model had the highest accuracy of settlement prediction when the mesh size of the finite element model was 1.4[Formula: see text]m, and the GA-SVR model had the highest accuracy and generalization ability in ground settlement prediction. This study highlights the uniqueness of machine learning–finite element mesh optimization model in application.

Development of a simple test to quantify the consolidation properties of liquefied granular materials

AbstractLiquefaction can have devastating consequences by causing increased mobility of debris flows, tailings dam breaches, and settlement following seismic shaking. Observations on the consolidation behaviour of liquefied soils in 1-g or centrifuge shake table tests have permitted significant advancements in analytical and numerical methods to predict the rate and magnitude of consolidation settlement. However, advanced consolidation models introduce material parameters which are currently difficult to define quickly and at low cost. The objective of this paper is to demonstrate the development of a simple, low-cost test that can be used to quickly estimate the spatially and temporally varying coefficient of consolidation, which controls post-liquefaction settlements. Using the proposed simple setup, experiments were conducted on four uniformly graded sands of varying grain size and on one well-graded mixture. A range of analytical and numerical approaches from literature were assessed for their ability to back-analyse the observed pore pressure dissipation and settlement. Estimated values for the coefficient of consolidation were comparable between models and estimates of surface settlement matched well the experimental results. The simplicity of the proposed test combined with its low-cost and mobile nature, raise significant possibilities for quick estimations of the evolution of the coefficient of consolidation post-liquefaction.

An aftermath analysis of caving characteristics and movement of overlying strata in fully mechanized longwall gob

The large-scale collapse of overlying strata in the gob directly affect the safe production of coal mines; they are also the major causes of geological disasters, such as ground cracks, surface subsidence, and ground collapse. In this paper, the movement and caved characteristics of overlying strata during coal seam excavation are studied by conducting a physical model experiment. Results show that overlying strata have different movement and caved laws during the initial, intermediate, and later mining stages. During the initial mining stage, overlying strata do not collapse, and the subsidence is extremely small. During the intermediate mining stage, overlying strata cave along the vertical direction, and caved height gradually increases. Large numbers of cavities, abscission layers, and fractures exist between caved strata. The fractured area gradually increases upward, and the subsidence increases considerably. During the later mining stage, overlying strata cave along the horizontal direction. The abscission layers between the caved strata of the central are compacted. The compacted area is surrounded by a fractured area. The compacted and fractured areas increase along the horizontal direction. The subsidence curves exhibit a horizontal variation. Overlying strata evolve from the self-equilibrium stage to the vertical collapse stage, and finally, the horizontal collapse stage. The fractured area changes from a no fractured area to a fractured area, increases vertically, and finally, increases horizontally. The subsidence curve changes from extremely small to large, and finally, changes horizontally.

The Longitudinal Push-Out Effect and Differential Settlement Control Measures of the Transition Section of Road and Bridge Induced by Freeze–Thaw Inducing

The environmental influence of seasonal freezing and thawing forces the longitudinal shear effect of the bridge abutment, and the differential settlement between the subgrade and the bridge abutment will significantly affect traffic safety. In this work, based on the finite element simulation analysis method, the longitudinal push-out effect and differential settlement of the transition section caused by cycles are systematically investigated, and the treatment results under different control measures (buffer layer thickness) are compared and analyzed. The results show that changing the thickness of the buffer material in the transition section has no significant influence on the overall temperature field of the subsurface. The longitudinal displacement of the transition region will be obvious under the condition of seasonal cycle, and its longitudinal thrust effect on the abutment shows a typical periodic law with the seasonal change. As the depth of the lower soil layer from the surface increases, the pushing effect becomes weaker and weaker. The development of the different subsoil settlements in the transition section also showed periodic changes with the passage of seasons. The differential settlement of the transition section after the buffer layer treatment can be effectively controlled, and the maximum value of the surface settlement of the roadbed after the 5 cm thick buffer material is reduced by 35%, compared with the two deformations of frostshocked bridges, where differential settlement after the buffer material treatment creates only tip deformation. After using a 15 cm thick buffer layer material treatment, the maximum settlement value of the surface settlement of the road base is reduced from 0.2 m to 0.01 m, which will not affect safety and driving comfort. The research conclusions can provide a reference for the design of road and bridge transition sections in frozen areas.

Improved Mobilized Strength Design Method for Multi-Support Excavation Deformation Analysis

The safe and reliable design of underground spaces ensures the safety of a structure itself and its surroundings. The traditional Mobilized Strength Design (MSD) method for a multi-support excavation deformation analysis ignores the effects of soil parameters and excavation boundary conditions. Therefore, to compensate for the shortcomings of the existing MSD method, this paper proposes an improved mobilized strength design (IMSD) method for a multi-support excavation deformation analysis. The improved incremental deformation mechanism further considers the effect of the soil friction angle, and the effect of excavation depth and the first support on deformation energy are also considered. Further, the excavation calculation process based on the IMSD method is given, and the effects of different calculation parameters on the IMSD solution of excavation deformation are discussed. The results show that the IMSD method can effectively consider the effect of boundary conditions and the excavated process on the excavation deformation. The traditional MSD method underestimates the excavation deformation and surface settlement by an average of 15–23%, while the IMSD solution is more consistent with the measured values. The study results can provide a theoretical reference for the design of multi-support excavation.

Deformation prediction of overlying strata in multi-seam mining based on the influence function method-key stratum hybrid model

The extraction of coal seams from subterranean strata has the potential to induce displacement and deformation within the overlying strata. This leads to fractures in the strata and surface subsidence. The repetitive mining of coal seams, especially in scenarios involving multiple coal seams, can reactivate previously stabilized rock strata, causing them to subside and deform again. This exacerbates deformation and compromises the structural integrity of the strata. Consequently, there is an imperative need to thoroughly investigate the patterns of movement exhibited by overlying rock layers as a result of the recurrent extraction of coal seams. In recent years, significant progress in this field has been the effective utilization of the Influence Function Method-Key Stratum (IFM-KS) hybrid model. This model, distinguished by its amalgamation of mechanical and geometric methodologies, has proven to be efficacious in scrutinizing the movement dynamics of overlying strata in the context of single-layer mining operations. Based on that, this paper introduces a novel model for predicting rock layer movement and deformation in multi-seam mining. The calculation results derived from this model reveal the dynamics of rock layers' movement and deformation following secondary disturbances during repetitive mining. The accuracy and reliability of this newly proposed model in predicting rock movement during dual-layer coal mining are validated through both theoretical calculations and UDEC numerical simulations. The results demonstrate a high degree of consistency between the numerically simulated rock displacement and mathematical calculations and the rationality of the model. Based on the rock strata movement calculation results, the corresponding porosity distribution is also derived. Compared to single-layer coal mining, multi-seam mining could cause greater changes in porosity in the goaf area and extensive damage to rock strata. More fracture pathways between coal seams occur and a volume of air-leakage quantity is expected. This computational model provides invaluable insights for predicting rock strata deformation and porosity distribution in repetitive mining scenarios of a similar nature.

Progressive development of soil arching and deformations in two- and three-dimensional trapdoor tests

In this study, two- and three-dimensional trapdoor tests were conducted using a newly developed trapdoor test device, to investigate the evolution of soil arching and fill deformation with different relative fill densities and fill heights. The test results show that the trapdoor settlements normalised by trapdoor width required for the soil arching ratios to reach minimum and ultimate values increased with fill height. In tests with dense fill, both the minimum and ultimate soil arching ratios were smaller compared to tests with loose fill at the same fill height. Two symmetrical slip surfaces developed from the trapdoor edges, with their inclination angles dependent on the dilatation angle of the sand. The proposed Gaussian distribution function with fitting parameters well matched the measured fill settlement profile. In the dense fill tests, the settlement trough’s width and volume loss induced by trapdoor settlement gradually decreased with fill elevation due to its strong shear dilation behaviour. Conversely, the loose fill exhibited weaker shear dilation behaviour, resulting in a larger settlement region. The settlement trough width and volume loss ratio, derived from the measured fill deformation, can be used to predict the fill deformation, including the surface settlement.

A prediction model for surface settlement during the construction of variable cross-section tunnels under existing structures based on stochastic medium theory

When predicting ground surface settlement caused by constructing a new tunnel beneath existing structures, traditional stochastic medium theory inadequately accounts for the effects of abrupt changes in the cross-sectional areas of variable-section tunnels and the presence of existing structures, potentially leading to significant errors. In this paper, an enhanced ground surface settlement prediction model, based on stochastic medium theory, is proposed. The model equates the settlement of the existing structure’s base slab to that of the overlying soil and divides the horseshoe-shaped tunnel cross-section into eight arc segments, which are calculated using a polar coordinate system. Furthermore, the model treats soil loss at the junctions of variable-section tunnels as a linear transition, introduces the concept of a linear transition segment for variable sections, and accounts for the superimposed effects of closely spaced twin-tunnel excavations. Based on this model, a general-purpose calculation program has been developed that enables the rapid prediction of ground surface deformation caused by a variable-section tunnel passing beneath existing structures, simply by inputting engineering parameters. Finally, the accuracy of the prediction model was validated through comparisons with field measurement data, finite element analysis results, and calculations based on traditional stochastic medium theory. The results indicate that the proposed prediction model demonstrates high consistency with field data and finite element analysis results, whereas traditional stochastic medium theory results exhibit significant errors. This model is scientifically valid and provides a reliable reference for predicting ground surface settlement in comparable variable-section metro tunnel construction projects.

Model test study on instability mechanism during shield under-crossing existing pipeline

This paper aims at investigating the instability mechanism that occurs during the under-crossing of an existing pipeline using model tests and numerical simulations. The strain in the pipeline, earth pressure, surface settlement, and mechanisms of stratum instability during the process of instability have been obtained. The research findings reveal three distinct growth stages in the longitudinal deformation of the existing pipeline, with the strain growth rate during the rapid growth phase being 2–3 times higher than that observed during the slow growth phase. The pipeline displays a deformation pattern with tensile stress in the upper section and compressive stress in the lower section, due to the uneven force distribution along its length. In the absence of pipelines, the distribution of surface collapse exhibits a distinct “V” shaped. As the clear distance between the existing pipeline and the tunnel (CDPT) increases, there is a transition from a “partial W” to a “positive W” type distribution. The presence of the pipeline exerts the most significant inhibiting effect on the soil directly above it. In cases where there is no pipeline or the CDPT is small, soil leakage primarily occurs in front of the excavation. As the CDPT increases, soil leakage occurs around the pipeline and the instability range of the stratum significantly expands. The stable strata above the pipeline gradually widen, and the area with a significant influence range shifts from the right side to the left side of the existing pipeline.

Research on the application of a hybrid intelligent algorithm in reducing the negative response of shield tunneling construction

Traditional shield project control relies on manual monitoring and management, which suffers from poor reliability, lag, and other defects. In this paper, a hybrid algorithm based on Bayesian optimization (BO), light gradient enhancer (LGBM), and nondominated sorting genetic algorithm-III (NSGA-III) is proposed to realize intelligent control of shield construction safety. The BO-LGBM-NSGA-III is a multiobjective optimization algorithm that takes cutterhead wear and shield energy consumption as objective functions and the nonlinear mapping function of surface settlement and construction parameters as fitness functions. Based on the Pareto distribution of the optimized backing structure construction parameters, reasonable limits of the control parameters of the shield tunneling machine are obtained. The conclusions are as follows. (1) The R2 of the test set for surface subsidence prediction is 0.916, and the RMSE is 2.655. (2) The proposed method can achieve multiobjective optimization, and the surface settlement, cutter head wear, and shield energy consumption are optimized by 49.92%, 24.38%, and 41.88%, respectively. (3) By using this method, both prediction and optimization can be realized, and the optimal control scheme can be obtained. The negative impact of shield tunneling construction has been significantly reduced.

Model test of foundation pit excavation adjacent to the existing double-lane tunnel

Abstract To study the influence of foundation pit excavation on the adjacent existing double-track tunnel, a model box test with a similarity ratio of 1/50 was carried out. In this paper, the mechanical parameters of loess around the tunnel are determined through laboratory tests, and the miscellaneous fill, loess, sand, tunnel structure, support structure, and retaining wall are prepared by similar principles. The variation of the surface settlement, the deformation of the retaining wall, and the surrounding rock and soil pressure of the tunnel were analyzed by using equipment such as the miniature earth pressure box, dial gauge, and strain gauge. The results of the study showed that the existence of the supporting structure makes the bending moment of the retaining wall show the phenomenon of negative value first, then positive value, and finally negative value. In the process of foundation pit excavation, when the horizontal distance of the model surface is 4 times the excavation depth, the foundation pit excavation has little effect on the settlement of the model surface. The earth pressure value of the surrounding rock decreases with the increase of the excavation depth of the foundation pit. The earth pressure value of the surrounding rock is the right > the left > the lower part > the upper part and the excavation of the foundation pit has the greatest influence on the upper and left sides of the surrounding rock close to the side of the foundation pit, and the least impact on the right side of the surrounding rock.

Spatiotemporal Evolution Analysis of Surface Deformation on the Beihei Highway Based on Multi-Source Remote Sensing Data

Under the interference of climate warming and human engineering activities, the degradation of permafrost causes the frequent occurrence of geological disasters such as uneven foundation settlement and landslides, which brings great challenges to the construction and operational safety of road projects. In this paper, the spatial and temporal evolution of surface deformations along the Beihei Highway was investigated by combining the SBAS-InSAR technique and the surface frost number model after considering the vegetation factor with multi-source remote sensing observation data. After comprehensively considering factors such as climate change, permafrost degradation, anthropogenic disturbance, and vegetation disturbance, the surface uneven settlement and landslide processes were analyzed in conjunction with site surveys and ground data. The results show that the average deformation rate is approximately −16 mm/a over the 22 km section of the study area. The rate of surface deformation on the pavement is related to topography, and the rate of surface subsidence on the pavement is more pronounced in areas with high topographic relief and a sunny aspect. Permafrost along the roads in the study area showed an insignificant degradation trend, and at landslides with large surface deformation, permafrost showed a significant degradation trend. Meteorological monitoring data indicate that the annual minimum mean temperature in the study area is increasing rapidly at a rate of 1.266 °C/10a during the last 40 years. The occurrence of landslides is associated with precipitation and freeze–thaw cycles. There are interactions between permafrost degradation, landslides, and vegetation degradation, and permafrost and vegetation are important influences on uneven surface settlement. Focusing on the spatial and temporal evolution process of surface deformation in the permafrost zone can help to deeply understand the mechanism of climate change impact on road hazards in the permafrost zone.

A new model for predicting surface subsidence of twin salt cavern gas storages with different shapes

A new model for predicting surface subsidence of twin salt cavern gas storages with different shapes

Data-Supported Prediction of Surface Settlement Behavior on Opencast Mine Dumps Using Satellite-Based Radar Interferometry Observations

To ensure the safe repurposing of post-mining landscapes, understanding and managing geotechnical risks, particularly ground movements such as settlements on opencast mining dump surfaces, is critical. Satellite-based radar interferometry (InSAR) technology offers highly detailed data on vertical ground movements with a high spatial and temporal resolution. By combining a data-driven approach, using InSAR-generated high-resolution datasets, with model-driven methods such as inverse modeling and classic time–settlement models, the efficient monitoring and prediction of opencast mine dump settlements can be achieved. This dual approach—leveraging advanced data analysis tools and precise modeling—yields valuable insights into spatial settlement behavior. In particular, classic time–settlement models are applied to the InSAR data through least square regression and Taylor approximation. The integration of both approaches enables the more robust, data-validated forecasts of key geotechnical indicators, such as the time to settlement stabilization and the expected maximum settlement over large areas. An application at a mine in central Germany illustrates the method by generating spatial predictions of the settlement behavior over more than 200 ha. In general, the results provide a comprehensive dataset for investigating other factors influencing the settlement behavior of opencast mine dumps.