Settlement Prediction Research Articles

Predicting and validating the load-settlement behavior of large-scale geosynthetic-reinforced soil abutments using hybrid intelligent modeling

Settlement prediction of geosynthetic-reinforced soil (GRS) abutments under service loading conditions is an arduous and challenging task for practicing geotechnical/civil engineers. Hence, in this paper, a novel hybrid artificial intelligence (AI)-based model was developed by the combination of artificial neural network (ANN) and Harris hawks’ optimisation (HHO), that is, ANN-HHO, to predict the settlement of the GRS abutments. Five other robust intelligent models such as support vector regression (SVR), Gaussian process regression (GPR), relevance vector machine (RVM), sequential minimal optimisation regression (SMOR), and least-median square regression (LMSR) were constructed and compared to the ANN-HHO model. The predictive strength, relalibility and robustness of the model were evaluated based on rigorous statistical testing, ranking criteria, multi-criteria approach, uncertainity analysis and sensitivity analysis (SA). Moreover, the predictive veracity of the model was also substantiated against several large-scale independent experimental studies on GRS abutments reported in the scientific literature. The acquired findings demonstrated that the ANN-HHO model predicted the settlement of GRS abutments with reasonable accuracy and yielded superior performance in comparison to counterpart models. Therefore, it becomes one of predictive tools employed by geotechnical/civil engineers in preliminary decision-making when investigating the in-service performance of GRS abutments. Finally, the model has been converted into a simple mathematical formulation for easy hand calculations, and it is proved cost-effective and less time-consuming in comparison to experimental tests and numerical simulations.

A hybrid data-driven model for geotechnical reliability analysis

• A hybrid data-driven model is proposed to quantify risk under uncertain parameters • during a tunnel's excavation process. • A real case is used to evaluate the effectiveness and feasibility of our proposed • approach. • A Markov-chain-based importance sampling is used to analyze settlement reliability. Tunnel boring machines are widely used to construct underground rail networks in urban areas. However, ground settlement due to complex geological conditions is an ever-present reality requiring continuous monitoring and management of risks. This paper addresses the following research question: How can we predict tunnel-induced ground settlement with engineering parameters, improve its predictive ability, and quantify its risks under uncertain parameters in complex geological conditions ? To this end, we develop a hybrid data-driven model that considers prior domain knowledge to effectively and accurately quantify risk under uncertain parameters during a tunnel's excavation process. Our model comprises: (1) a deep neural network (DNN) to construct a ground settlement prediction model; (2) the incorporation of physical knowledge into the DNN-based prediction model; and (3) a Markov-chain-based importance sampling to analyze settlement reliability. We use the San-yang Road tunnel project in Wuhan, China, to evaluate the effectiveness and feasibility of our proposed approach. The results demonstrate that our hybrid data-driven model can accurately predict tunnel-induced ground settlement and quantify failure probability for geotechnical reliability under uncertain parameters during a tunnel's excavation process.

Predicting settlement of embankments built on PVD-improved soil using Bayesian back analysis and elasto-viscoplastic modelling

In this study, an implicit finite difference program, 1DFDM, based on the fully coupled one dimensional elasto-viscoplastic (1D EVP) model combining radial and vertical drainage is developed to perform consolidation analysis of embankments constructed on soft soils. Class A prediction (prior prediction) is performed by using 1DFDM for the embankment constructed on soft clay in Ballina, Australia. However, the prediction deviates significantly from the measurements, and the model parameters need to be calibrated. Bayesian back analysis is thus combined with 1DFDM in this study to do Class C prediction. Three updating schemes (using monitoring settlement only, using monitoring pore water pressure only, and using both monitoring settlement and pore water pressure) are employed to shed light on the effects of the monitoring data types on settlement prediction. The results show that the long-term settlement and pore water pressure predictions could be further improved if both the monitoring settlement and pore water pressure data are incorporated. The coupling effects in the constitutive model can reduce the fluctuations in the updated optimum parameters. A stable and physically reasonable parameter set could be obtained if enough monitoring data is incorporated into the Bayesian back analysis.

Prediction of long-term differential track settlement in a transition zone using an iterative approach

A methodology for the simulation of long-term differential track settlement, the development of voided sleepers leading to a redistribution of rail seat loads, and the evolving irregularity in vertical track geometry at a transition between two track forms, is presented. For a prescribed traffic load, the accumulated settlement is predicted using an iterative approach. It is based on a time-domain model of vertical dynamic vehicle–track interaction to calculate the contact forces between sleepers and ballast in the short-term. These are used in an empirical model to determine the long-term settlement of the ballast/subgrade below each sleeper. Gravity loads and state-dependent track conditions are accounted for, including a prescribed variation of non-linear stiffness of the supporting foundation along the track model. In parallel, a two-dimensional (2D) non-linear finite element model of layered soil is verified versus field measurements and used to determine the support stiffness of each sleeper in the track model. The methodology is applied to a transition zone between a ballasted track and a slab track that is subjected to heavy haul traffic. Analyses of the influence of higher axle loads and the implementation of under sleeper pads on sleeper settlement are demonstrated.

Closed-form consolidation solutions for known loading functions

In engineering practice, loading varies with time. However, the classical one-dimensional theory of consolidation assumes the stress increase is instantaneously applied. Many approaches to the problem of time-dependent loading have been proposed over the years, from approximate methods to full developments of differential equations. The paper presents a simple method for finding a closed-form consolidation solution for time-dependent loading without the need for differential equations. Two sets of general equations were derived for both excess pore pressure and average degree of consolidation. Equations were solved for linear, parabolic, sinusoidal, and exponential load functions. Stepped and cyclic loads were also addressed and a numerical solution was developed to verify the obtained result. The method proved to be easy to apply and provides solutions with great simplicity. A case study of non-instantaneous loading on soft clay was also analyzed, and settlement prediction showed good results when compared to readings of the settlement plates.

A novel probability integral method segmental modified model for subsidence prediction applicable to thick loose layer mining areas.

In response to the problem that the actual extent of coal mining impacts on the surface in thick loose layer mines significantly exceeds the theoretical predictions, based on the literature study, the form of influence of thick loose layer on the predicted parameters of the probability integral method is summarized and analyzed; taking into account the influence of the subsidence coefficient, the sine modification formula of the major influence radius and the logistic modification formula of the subsidence coefficient are established, respectively, and based on the characteristics of the major influence radius, a new subsidence basin demarcation point is proposed and a novel probability integral method segmental parameter modified prediction model is constructed. The simulated experiment and real data experiment results prove that the constructed probability integral method segmented parameter modified model can both reduce the convergence of surface subsidence basin edge better and take into account the predicted accuracy inside the subsidence basin. The research achievements provide scientific data support for disaster warning, pollution management, ecological restoration, and coordination between coal mining and surface city construction in thick loose layer mining areas.

Analysis of Settlement of Group Pile Foundation in Linear Viscoelastic Soil

The settlement of a group pile foundation is the result of elastic-plastic settlement of the pile and rheological settlement of the soil. Unfortunately, most research only considers elasto-plasticity settlement of piles or rheological settlement of soils. Classical theories like the Poynting-Thompson model only provide some prediction for the elasto-plasticity behavior of pile groups, which cannot satisfactorily reflect the total settlement characteristics of pile group foundation. In the current paper, we presented a method to study settlement behaviors of pile group foundation in linear viscoelastic soil based on the pile-pile interaction factors, the calculated equation of settlement of single piles and pile groups under vertical loads that are derived, the influence law of pile end resistance, pile spacing, slenderness ratio, elastic modulus ratio, and Poisson ratio on pile end load ratio, and the interaction factors that have been presented. Finally, some typical predictive examples have been presented for a 2 × 2, 3 × 3 group pile showing settlement from a field test. It is found that the pile-pile interaction factors increase with the increase of the pile-soil elastic modulus ratio and length-to-diameter ratio but decrease with the increase of the ratio of distance to diameter, and the ratio of distance to diameter has little effect on the interaction factor when pile spacing is more than 10 d (d is the diameter of the pile). The error between the field test results and computed data is 14.9%, the settlement calculated by the present method shows good agreement with the reported results, and the correlativity coefficients R2 are greater than 0.9223, which signifies that the present method is suitable for the theoretical calculation and prediction of the composite foundation settlement of linear viscoelastic soil when piles are used to strengthen the foundation. The method would provide a unique view to quantitative prediction of the settlement characteristic of group pile foundation.

Land subsidence modeling and assessment in the West Pearl River Delta from combined InSAR time series, land use and geological data

West Pearl River Delta (WPRD) is sinking as a result of the jointed effect of natural and anthropogenic factors. Land subsidence has increasingly become a concern because of tremendous population growth and rapid urbanization over this region in the last few decades. In this study, sixty-seven Sentinel-1 images, acquired between 2016 and 2021, were analyzed with the persistent scatterer interferometry technique (PSI), to monitor and reveal the ground subsidence characteristics in the WPRD. It is found that the overall vertical deformation velocities observed in the WPRD ranged between −70 mm/year and 10 mm/year. Three subsidence bowls were found in the study area (Gaolan island of Zhuhai, the junction area of Zhuhai and Zhongshan, and the junction area of Zhongshan and Jiangmen). The spatial–temporal subsidence characteristics have been analyzed. It is discovered that the ground subsidence is mostly dispersed in Quaternary deposits and is highly relevant to the thickness of sediments, indicating that soft soil consolidation is one of the primary causes contributing to land subsidence. Furthermore, land use maps for 2016 and 2021 were generated using Landsat-8 images for the investigation on the relationship between land subsidence and land use. The results obtained from analysis demonstrated that the rapid subsiding areas mainly occurred in the land-use classes as follows: aquaculture, urban land, and agricultural land. The land use conversion pattern with more significant anthropogenic influence usually causes a higher subsidence rate. In addition, based on soft soil thickness, groundwater exploitation, land use, elevation, and strata lithology, a Random Forest Regression (RFR) model was used to predict subsidence rates (R2 = 0.631, RMSE = 2.7 mm/year). The importance of these influencing factors of land subsidence was calculated based on the RFR algorithm. The results indicated that soft soil thickness, elevation, groundwater exploitation, strata lithology, and landcover type are the most significant factors affecting subsidence. The applicability of geological data and land-use history for land subsidence prediction has been demonstrated with the use of the RFR algorithm.

Hybrid Random Forest-Based Models for Earth Pressure Balance Tunneling-Induced Ground Settlement Prediction

Construction-induced ground settlement is a serious hazard in underground tunnel construction. Accurate ground settlement prediction has great significance in ensuring the surface building’s stability and human safety. To that end, 148 sets of data were collected from the Singapore Circle Line rail traffic project containing seven defining parameters to create a database for predicting ground settlement. These parameters are the tunnel depth (H), the tunnel advance rate (AR), the EPB earth pressure (EP), the mean SPTN value from the soil crown to the surface (Sm), the mean water content of the soil layer (MC), the mean modulus of elasticity of the soil layer (E), and the grout pressure used for injecting grout into the tail void (GP). Three hybrid models consisting of random forest (RF) and three types of meta-heuristics, Ant Lion Optimizier (ALO), Multi-Verse Optimizer (MVO), and Grasshopper Optimization Algorithm (GOA), were developed to predict ground settlement. Furthermore, the mean absolute error (MAE), the mean absolute percentage error (MAPE), the coefficient of determination (R2) and the root mean square error (RMSE) were used to assess predictive performance of the constructed models for predicting ground settlement. The evaluation results demonstrated that the GOA-RF with a population size of 10 has achieved the most outstanding predictive capability with the indices of MAE (Training set: 2.8224; Test set: 2.3507), MAPE (Training set: 40.5629; Test set: 38.5637), R2 (Training set: 0.9487; Test set: 0.9282), and RMSE (Training set: 4.93; Test set: 3.1576). Finally, the sensitivity analysis results indicated that MC, AR, Sm, and GP have a significant impact on ground settlement prediction based on the GOA-RF model.

Numerical inspection of Miner’s rule and drained cyclic preloading effects on fine-grained soils

The prediction of soil cyclic behaviour is relevant for several geotechnical applications. In order to obtain accurate predictions of settlements and pore water pressure build-up on boundary value simulations, robust constitutive models which are able to realistically reproduce the soil mechanical behaviour under monotonic and cyclic loading are necessary. To have confidence on their predictions, their capabilities and limitations should be very well known. In this article, the prediction capabilities and limitations of two advanced constitutive models for fine-grained soils, namely: CAM and AHP+ISA were inspected considering undrained cyclic triaxial tests with more complex and realistic nature than ”conventional” undrained cyclic tests usually considered in the literature for the calibration and validation of constitutive models but not representative of many real loading scenarios. In particular, the influence of drained cyclic preloadings and packages of cycles with different deviatoric stress amplitudes in different sequences (inspection of Miner’s rule) was evaluated. For that purpose, simulation results with the models were qualitatively as well as quantitatively compared against several experimental results under monotonic and cyclic loading on Malaysian kaolin reported by Duque et al. (2022a).

Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter

Bayesian model updating provides a powerful framework for updating and uncertainty quantification of models by making use of observations, following probability rules in the treatment of uncertainty. Particle filter (PF) and Bayesian Updating with Structural Reliability method (BUS) have been developed by researchers as promising computational tools for this purpose. However, reducing computational cost in the updating process, especially for complex models, remains one of the key challenges. Surrogate model approach achieves this by appropriately replacing, possibly adaptively, the evaluation of the original computationally costly models with approximate ones that are much less costly. This study proposes an efficient method to estimate the posterior probability density function (PDF) of model parameters by using a surrogate model constructed using adaptive Gaussian Process Regression and PF. Of critical importance is the development of ‘learning function’, which finds the location of large values of posterior PDF and avoids those that have been visited. The proposed methodology is illustrated using a single-variable example and compared with PF and BUS. Its application is illustrated through an example of structural dynamics and another one on settlement prediction by soil–water coupled FEM with Cam-clay model.

Prediction of fine tailings settlement in pit lakes using a population growth model

A model for predicting the settlement trajectory in saturated fluid fine tailings deposits is described. The model uses a population growth function and leverages established theories in soil mechanics, clay–water surface interactions and biogeochemistry to derive compressibility and permeability functions for predicting the settlement behaviour of fine tailings in deep deposits, typical of end pit lakes. The method is particularly useful for tailings treated with flocculants and/or coagulants with continuously changing compressibility and permeability parameters during deposition. For oil sands or mineral sands fluid tailings treated with or without high doses of a coagulant and/or a flocculant, the consolidation parameters determined from the model are comparable with those measured using standardised large-strain consolidation methods, and the predicted settlement using a modified Gibson’s finite-strain equation closely describes the measured settlement and void ratio profiles in geocolumns.

Numerical Analysis of Settlement of a Piled Raft Foundation on Coastal Soil

There is a growing demand for multi-story buildings for residence and commercial purposes in coastal areas of Sindh, Pakistan. Such types of soils are generally considered more compressible with high groundwater levels, which may cause lower shear strength and higher settlement. The computation of the settlement of foundations requires the use of advanced constitutive models, which are not commonly used due to a lack of field or experimental data. This study is carried out to illustrate the use of an advanced soil model, i.e., Hardening Soil Model for the computation of settlement. For this purpose, numerical modeling was carried out using Finite Element Program PLAXIS 2D. Initially, the MC Model was utilized for the calculation of the settlement of a 10-story building in the coastal soil. In addition, parametric analyses for the effects of modulus of elasticity, permeability, and dilatancy angle were carried out. The results mainly suggest that the settlement of the building constructed on a piled raft foundation, predicted with the MC model, was 40% higher than that of the HS model. For prediction of settlement of the piled raft foundation, the results suggest that the HS model can be given preference as compared to the MC model. Doi: 10.28991/CEJ-2023-09-02-05 Full Text: PDF

Experimental research on the influence of different factors on the behaviour of broken coal and rock particles during compaction

Broken coal-rock masses are a major part of the goaf. The breaking of coal and rock particles in the compaction process affects various aspects including control of strata motion, prediction of surface subsidence, and backfill mining. Therefore, isotropic compaction experiments on broken coal-rock mass were carried out by utilising the loading part of the seepage test system for coal-rock mass to analyse influences of particle strength, particle size, mixed type, stress, and loading and unloading times on breakage of particles in the compaction process. The test results show that the crushing behaviour of coal and rock particles in the compaction process can be divided into three stages: the structural re-arrangement and crushing of particles under a low stress, particle breakage under moderate stress, and compression of particles under a high stress. The reduction rate of porosity decreases with the increase of particle strength and increase of loading and unloading times, increases with the increase of particle size, has nothing to do with two mix types of coal-rock mass after compaction. Under the same stress unloading conditions, the recovery amounts of particles from compression deformation were similar, which was not related to unloading times. Research results is significant to those analysing overlying strata motion, predicting surface subsidence, optimising the material ratio for backfill mining, etc.

Spatial-temporal fusion network for maximum ground surface settlement prediction during tunnel excavation

The maximum ground surface settlement prediction is a complex problem as the settlement depends on plenty of intrinsic and extrinsic factors. To obtain the approximate range of the settlement, a hybrid prediction dataset including the geological and construction parameters is built using spatial and temporal series according to the sampling methods. The settlement prediction task is transformed into a multi-modal and multi-variate series prediction task. Hence, a spatial-temporal fusion network (STF-Network) is proposed. The spatial-temporal fusion mechanism is firstly designed to establish the spatial-temporal fusion map, which makes spatial and temporal series interact earlier. Then, the 3D residual unit structure is designed to capture the features of temporal series and spatial-temporal fusion map, and two fully-connected layers are established to capture the spatial structural information. Finally, the final output is merged by the three components. The experimental results for STF-Network demonstrate the superiority over state-of-the-art methods.

Comparative Study on Ground Improved by Stone Columns and Ground Improved by Hybrid Concrete Columns

This paper provides a detailed discussion and comparison of two deep ground improvement techniques, stone columns and hybrid concrete columns (HCC’s), in terms of design, construction procedure and performance, through a case history from a recent petrochemical project in Malaysia. Two full-scale group load tests (GLT’s) were conducted on test columns constructed prior to construction of production columns and used to calibrate the analytical models adopted for the design of stone columns and HCC’s. The ground improved by HCC’s shows a stiffer load-settlement behavior compared to the ground improved by stone columns as predicted by analytical models. In addition, the time-dependent settlement is more obvious in the stone column improved ground than in the HCC improved ground. Stone columns may act as a drainage path helping expulsion of pore water as the applied load increases whereas concrete columns transfer the load to the surrounding soils through the load transfer at the interface between a concrete column and the surrounding soil. The results of all four GLT’s on production columns confirm that the settlement predictions with the calibrated analytical models are satisfactory. This paper also summarizes the key calibration techniques adopted at the project site for quality control during construction. These calibration techniques can be considered as a useful guide for other project sites with similar ground improvement methods proposed in similar ground conditions.

Research on Ground Surface Settlement of Curved Pipe Jacking Based on Numerical Simulation

The construction disturbance caused by pipe jacking is an index that must be strictly controlled in the project. With the wide application of curved pipe jacking in urban underground construction, predicting its ground surface settlement characteristics has become an increasingly important research topic. Based on finite element numerical simulation, a three-dimensional model of curved pipe jacking was established to simulate the construction stage. Compared with the straight pipe jacking, the normal component of jacking force is set in the outer tunnel of the curved pipe jacking to simulate the extrusion of the outer soil caused by the corner of the pipe joint, and the surface settlement characteristics of the curved pipe jacking are explored by changing the pressure value. The research results show that the surface settlement of the curved pipe jacking conforms to the normal distribution as a whole. Compared with the linear pipe jacking, the surface settlement is not symmetrical about the design axis of the tunnel. With the increase of the pressure outside the tunnel, the maximum settlement value of the surface and the difference between the two sides of the axis increase. Compared with the Peck formula, the width of the settling tank of the curved pipe jacking is larger, the convergence rate outside the width of the settling tank is slower, and the settlement trend is more in line with the Verruijt formula. The calculation formula is modified by axis offset distance and ovalization coefficient and modified formula can better reflect the measured engineering data. The research results can provide guidance and reference for the prediction and control of surface settlement of curved pipe jacking.

Isogeometric Analysis of Longitudinal Displacement of a Simplified Tunnel Model Based on Elastic Foundation Beam

Serious uneven settlement of the tunnel may directly cause safety problems. At this stage, the deformation of the tunnel is predicted and analyzed mainly by numerical simulation, while the commonly used finite element method (FEM) uses low-order continuous elements. Therefore, the accuracy of tunnel settlement prediction is not enough. In this paper, a method is proposed to study the vertical deformation of the tunnel by using the combination of isogeometric analysis (IGA) and Bézier extraction operator. Compared with the traditional IGA method, this method can be easily integrated into the existing FEM framework, and ensure the same accuracy. A numerical example of an elastic foundation beam subjected to uniformly distributed load and an engineering example of an equivalent elastic foundation beam of the tunnel are given. The results show that the solution of the IGA method is closer to the theoretical solution of the initial-parameter method than the FEM, and the accuracy and reliability of the proposed model are verified. Moreover, it not only provides some theoretical support for the longitudinal design of the tunnel, but also provides a new way for the application and popularization of IGA in tunnel engineering.

Research on Settlement Prediction of Building Foundation in Smart City Based on BP Network

Research on Settlement Prediction of Building Foundation in Smart City Based on BP Network

Consolidation settlement prediction and monitoring of toll road embankment at STA 23+650 Semarang–Demak Toll Road section

The development of toll road projects in Indonesia is experiencing rapid growth. However, due to the large number of toll roads being constructed over soft soil areas, it is crucial to address the issue of consolidation settlement. This is because settlement can lead to differential or excessive settlement, which potentially damages the structures built on such soil. Unfortunately, consolidation settlement on soft soils is a time-consuming process. Thus, a consolidation acceleration method is needed, one of which is to use a combination of PVD–Preloading. This method can also increase the soil’s bearing capacity, thereby contributing to the preservation and protection of systems that support human activities while maintaining a balance with nature. Therefore, this study aims to investigate consolidation settlement employing PVD–Preloading methods. The analysis focused on a case study conducted at STA 23+650 of the Semarang-Demak Toll Road section. The analysis employed the Asaoka method, integrating field monitoring instruments, along with the Terzaghi method for one-dimensional consolidation. The results showed that the settlement results from the Asaoka and Terzaghi methods correlate with the settlement plate instrumentation in the field.