Back Analysis Method Research Articles

Back analysis of geomechanical parameters based on a data augmentation algorithm and machine learning technique

Accurate geomechanical parameters are key factors for stability evaluation, disaster forecasting, structural design, and supporting optimization. The intelligent back analysis method based on the monitored information is widely recognized as the most efficient and cost-effective technique for inverting parameters. To address the low accuracy of measured data, and the scarcity of comprehensive datasets, this study proposes an innovative back analysis framework tailored for small sample sizes. We introduce a multi-faceted back analysis approach that combines data augmentation with advanced optimization and machine learning techniques. The auxiliary classifier generative adversarial network (ACGAN)-based data augmentation algorithm is first employed to generate synthetic yet realistic samples that adhere to the underlying probability distribution of the original data, thereby expanding the dataset and mitigating the impact of small sample sizes. Subsequently, we harness the power of optimized particle swarm optimization (OPSO) integrated with support vector machine (SVM) to mine the intricate nonlinear relationships between input and output variables. Then, relying on a case study, the validity of the augmented data and the performance of the developed OPSO-SVM algorithms based on two different sample sizes are studied. Results show that the new datasets generated by ACGAN almost coincide with the actual monitored convergences, exhibiting a correlation coefficient exceeding 0.86. Furthermore, the superiority of the OPSO-SVM algorithm is also demonstrated by comparing the displacement prediction capability of various algorithms through four indices. It is also indicated that the relative error of the predicted displacement values reduces from almost 20% to 5% for the OPSO-SVM model trained with 25 samples and that trained with 625 samples. Finally, the inversed parameters and corresponding convergences predicted by the two OPSO-SVM models trained with different samples are discussed, indicating the feasibility of the combination application of ACGAN and OPSO-SVM in back analysis of geomechanical parameters. This endeavor not only facilitates the progression of underground engineering analysis in scenarios with limited data, but also serves as a pivotal reference for both researchers and practitioners alike.

Enhancing ground loss rate prediction in soft-soil shield tunneling: a synergistic approach of peck back analysis and eXtreme Gradient Boosting and bayesian optimization

This study investigates the prediction of ground loss rate during soft-soil shield tunneling using Peck’s back analysis method and XGBoost model. Bayesian optimization is employed to determine optimal hyperparameters, ensuring comprehensive and efficient model tuning. The XGBoost model is compared with Random Forest (RF) and Support Vector Machine (SVM) models to benchmark its performance. The results demonstrate the superior accuracy and robustness of the XGBoost model. Also, the results show that the soil properties and the grouting factors of the excavation face affect the duration of the instantaneous settlement of the ground surface. There is a specific correlation between the depth-to-diameter ratio, the coefficient of variation in the advancing speed of the shield machine, the maximum surface subsidence, and the ground loss rate. The prediction model of the ground loss rate based on the combined approach of Peck back analysis and eXtreme Gradient Boosting and Bayesian optimization has high reliability in soft-soil layers, and this method can provide a specific reference for predicting construction risk in related projects.

Simulation of geological uncertainty based on improved three-dimensional coupled Markov chain model

The natural stratigraphic distribution is uncertain due to the geological tectonic movements and sedimentation. It is a challenge for traditional deterministic models to capture geological uncertainty. The objective of this study is to develop an improved three-dimensional coupled Markov chain (ICMC3D) model based on limited boreholes. The proposed method enhances the traditional three-dimensional coupled Markov chain model by modifying the transition probability equation for predicting unknown elements. The dip data is added to estimate the transition probability matrix with the borehole information. On this basis, a back analysis method is proposed for estimating the horizontal transition probability matrics. The method makes the predicted results more suitable for the possible stratigraphic distribution. The accuracy of the proposed method is verified through some applications. In this paper, an improved three-dimensional coupled Markov chain model and transition probability matrix estimation method are illustrated by the borehole data collected from a tunnel project in Qingdao. The results show that the presented method can reflect the asymmetry, continuity and anisotropy of the three-dimensional stratum. The exceedance probability is proposed to predict the fault zone development range in order to reduce the geological uncertainty.

Seroja Tropical Cyclon Destruction Case Study: Permanent Slope Protection Evaluation at Downstream of Rotiklot Dam Nomenclature Building, East Nusa Tenggara, Indonesia

Abstract Rotiklot Dam was one of some infrastructures which was affected by the Seroja Tropical Cyclone in East Nusa Tenggara Indonesia in April 2021. There are several landslides at the downstream of the dam that can threaten the safety of the dam, especially at the downstream of the nomenclature building. The right design of permanent slope protection must be determined in a limited time. Geotechnical investigations such as boring, resistivity survey, standard penetration test, and water pressure test were conducted in a short time. Back analysis method was carried out to find extreme soil parameters conditions. The slope protection must meet the national standard minimum safety factor or greater than 1.5. Safety factors of slope protections are evaluated by a finite element method to get the safety factor and maximum total displacements. Every stage of slope protection can give the optimal safety factor values which are required or greater than 1.5, however backfill material will give more load and decrease the safety factor value.

Back analysis of the shear strength parameters of 3D landslides with Bayesian approach and reliability theory

To establish an accurate model for determining the shear strength parameters of sliding surfaces while considering the variability and uncertainty of geotechnical parameters, a reliability back analysis method for parameters based on Bayesian theory and a three-dimensional slope stability analysis method are proposed in this paper. First, the prior information of shear strength parameters is updated according to the field test information based on Bayesian theory. Then, combined with the reliability theory, the influences of different two-dimensional and three-dimensional sliding surface shapes on the parameters back analysis are studied. The results show that the uncertainty of parameters decreases with the increase in the amount of test data after introducing the test data to update the mastered parameter information based on Bayesian theory, and it is of great significance to consider the shape and three-dimensional effect of the sliding surface accurately to improve the accuracy of parameter determination and slope stability analysis.

Bayesian Back Analysis of Axially Loaded Pile Response Using Fibre Optic Strain Data

The load-transfer method has been extensively used to evaluate the deformation behavior of foundation piles, with the nonlinear load-settlement behavior of a pile-soil system modeled using load-transfer curves. Although load-transfer models have a considerable impact on the pile response, the model parameters are usually estimated empirically from soil parameters using transformation models, and the associated uncertainties are often neglected. Fiber optic sensing technology provides detailed axial strain information along the entire length of a pile; however, the full potential of such utilization has not yet been realized. This paper presents a Bayesian back analysis method for statistical descriptions of load-transfer curves and pile responses based on fiber optic measurement. A pile test case study is used to illustrate the proposed method, and several candidate model configurations are compared. The results show that the predicted pile response generally matches the measured data, and the quantified uncertainty allows for a better understanding of the possible ranges of the pile response, in addition to the single best-estimate result.

An artificial intelligence optimization method of back analysis of unsteady-steady seepage field for the dam site under complex geological condition

An artificial intelligence optimization method of back analysis of unsteady-steady seepage field for the dam site under complex geological condition

Back analysis of shear strength parameters of slope based on BP neural network and genetic algorithm

AbstractEfficient and accurate acquisition of slope shear strength parameters is the key to slope stability analysis and landslide prevention engineering design. This paper establishes a back analysis method based on uniform design, artificial neural network, and genetic algorithm. It can obtain the shear strength parameters of slopes based on information such as the radius and center coordinates of the slip surface obtained from on‐site investigations. This method has been applied to engineering practice. The research results indicate that the stability of the waste dump slope is most sensitive to the response of the internal friction angle of the loose body, followed by cohesion, and least sensitive to the response of the soil volume weight. This method can effectively reduce the number of network training samples and efficiently and quickly determine the initial weights of the BP (abbreviation for back‐propagation) neural network. This method can efficiently and quickly conduct back analysis to obtain the shear strength parameters of slopes. Using the obtained shear strength parameters for slope stability calculation, the most dangerous slip surface abscissa error, ordinate error, and slip surface radius error are only 3.59%, 0.95%, and 1.83%. It is recommended to promote the back analysis method of shear strength parameters in engineering practice in the future.

Back analysis of cohesion and friction angle of failed slopes using probabilistic approach: two case studies

In general, geotechnical investigations deal with the back analysis of the geotechnical parameters on the basis of the field observation. According to the back analysis method introduced in the present study, geometry of a number of failed slopes have been carefully mapped and then statistical features of the groundwater level, the bulk unit weight and other measurable parameters have been determined. After that, a series of two-dimensional models for back analysis of the failures have been established. Moreover, statistical analyses based on probabilistic approaches have been utilized to estimate the variation ranges of the shear strength variables. The study provided the probabilistic back analysis of the slope failure with the two case studies in the copper mines of Anjerd and Daraloo. Results indicated that the approach to the probabilistic back analysis has been effective in the analysis of the slope failures wherein considerable uncertainty has been observed in the shear strength and other influential variables. Furthermore, the probabilistic back analysis presented a lot of information in comparison to the approach to the deterministic back analysis and thus had more reasonable matching with the practice of geotechnical engineering in the real world. Therefore, this method would be beneficial and practical for stability analysis, redesigning the slopes, designing the new slopes under the same geotechnical conditions and promote the construction safety. The probabilistic back analysis could be also used to estimate the shear parameters and analyze the stability of slopes as a cost-effective and high reliability method.

The Mechanism of Rainfall-Induced Landslide Around Railway Tracks in Central Java Province, Indonesia

Landslide is one of the most disastrous natural hazards in Indonesia, causing significant fatalities and economic losses. Landslides can be triggered by several factors, such as rainfall, earthquakes, soil conditions, and others, where each landslide event has its own triggering and controlling factors. A progressive landslide occurred on the Central Java railway line which resulted in damage to the double-track railway as a transportation infrastructure. The objective of this paper is to understand the process and triggering factors of the landslide. Information was collected through field investigations and measurements based on drilling results at 3 points, geophysical surveys at 5 lines, and laboratory testing of several soil samples. Geological and geotechnical settings, topography, lithology, hydrogeology, and rainfall data of the area were analyzed. Aerial photographs and other remote sensing data were used to evaluate and discuss the information. Landslides in the study area occurred in stages, starting with the formation of a tension crack, followed by two landslides over five months. The results show that the clay material that dominates the study area is the dominant controlling factor of a landslide, triggered by long-duration, low-intensity rainfall. Rainwater entering through tension cracks increases moisture content, adding load to the slope and triggering landslides. Furthermore, the train's external load on the slope also contributes to the occurrence of landslides. The static and cyclic load of the train causes changes in the slope's pressure balance, generating a force that drives the downslope soil. Further analysis was performed using back analysis method with the limit equilibrium method to enhance understanding of slope stability parameters at the time of slope failure. The analysis was performed considering the rising groundwater level. A factor of safety (FS) value of 0.989 was obtained at the end of the simulation, indicating that the slope had failed.

Review on the Estimation of Static Deformability Modulus of Rocks and their adoptability in Different Rock Masses

The aim of this study is to review the different mechanisms employed in the estimation of static rock mass deformability modulus (𝐸𝑚) in rock engineering applications and to investigate the adoptability of the identified mechanisms in different rock masses. The paper discusses different evaluation criteria through experimental, empirical and other means, with their merits and demerits, including influential factors. It is known that deformability modulus of intact rock depends on the imposed stress, strain rate and the confining stress on the rock sample as well as the rock texture and structure. The results generated for 𝐸𝑚 by different in-situ tests are different and an appropriate in-situ test based on the rock mass conditions should be employed to obtain reasonable results. Empirical criteria are found to produce results of reasonable precision if appropriately adopted for specific rock mass conditions, while the back analysis method is widely adopted as an insitu estimation measure for the design of rock-sockets and tunnel support. It has also been reported that substantial reduction in 𝐸𝑚 occurs due to schistosity and larger test volumes, while it is sensitive to stress and discontinuity conditions. In this work, specific recommendations are made on the estimation of 𝐸𝑚 for different types of rock masses based on the findings and reviews reported in the literature.

Back analysis of rock mass parameters in tunnel engineering using machine learning techniques

Efficient determination of the rock mass properties is vitally important for calculating and evaluating tunnel stability in tunnel engineering. The back analysis method has been widely used as an indirect method for determining rock mass parameters based on field measurements. However, most back analysis methods are generally time-consuming for numerical simulation and are merely based on the measured displacement, which leads to the identification of rock mass parameters that cannot fully reflect the characteristics of the surrounding rock. To improve the accuracy of the estimation of rock mass parameters, this paper presents a back analysis method based on multi-output support vector regression (MSVR) and differential evolution (DE) algorithms. Firstly, the global sensitivity analysis of rock mass parameters is analyzed using the elementary effects method. Numerical simulation is then carried out to prepare training samples. DE algorithm is used to determine the optimum hyperparameters of MSVR. Based on the monitoring data, the rock mass properties of the selected sensitive parameters are estimated by the constructed MSVR model. A high-speed railway tunnel is utilized to demonstrate the effectiveness of the MSVR with the DE algorithm (DE-MSVR). The results show that the DE-MSVR with mixed monitoring data of vault settlement, convergence, and rock mass stress has higher forecasting performance than these models with a single type of monitoring data. It is feasible to use the monitoring data at the early stages combined with the numerical simulation for parameter back analysis. Moreover, the comparison results show that the presented method exhibits higher prediction accuracy than the existing back analysis models.

Analysis on intelligent deformation prediction of deep foundation pits with internal support based on optical fiber monitoring and the HSS model

With the continuous expansion of the city scale, while rapidly advancing foundation pit construction, it has become a top priority to prevent the potential safety hazards caused by efforts to catch up with deadlines. In this case, deep foundation pit monitoring can provide important technical support for the foundation pit design and construction safety. However, conventional foundation pit monitoring for point monitoring of structural characteristics cannot accurately locate local strains and related cracks, and cannot provide real-time monitoring of support structures, ordinary soil constitutive models cannot consider the strain hardening characteristics of soft soil. This paper aimed to analyze the characteristics of excavation deformation of a deep foundation pit with internal support in the Dalian Donggang Business District by combining the methods of optical fiber monitoring and finite element simulation. In this study, distributed optical fiber monitoring and routine monitoring were adopted to carry out the synchronous excavation monitoring of foundation pit support structures. The main monitoring objects were the deep horizontal displacement of the support piles and the surface settlement of the foundation pit. Moreover, the authors used the Plaxis finite element software based on the Hardening Soil-small (HSS) model to conduct the numerical simulation analysis, and the results were compared with two groups of the measured data of the deep foundation pit obtained from the aforementioned research. Additionally, in light of the SSA-BP neural network, the back-analysis method of HSS model parameters in the Dalian area is proposed. The findings show that, under the premise of selecting reasonable parameters, the results of finite element analysis of the HSS model, considering the small strain characteristics of the soil, demonstrate a high degree of fit with the actual deformation law of foundation pits, which proves the rationality of the model. Furthermore, it was verified that the distributed optical fiber monitoring has conspicuous advantages in terms of data continuity and accuracy in contrast to routine monitoring. The research results of this paper can provide reference for deformation monitoring and early warning regarding the support structures of deep foundation pits.

A modified back analysis method for deep excavation with multi-objective optimization procedure

Real-time prediction of excavation-induced displacement of retaining pile during the deep excavation process is crucial for construction safety. This paper proposes a modified back analysis method with multi-objective optimization procedure, which enables a real-time prediction of horizontal displacement of retaining pile during construction. As opposed to the traditional stage-by-stage back analysis, time series monitoring data till the current excavation stage are utilized to form a multi-objective function. Then, the multi-objective particle swarm optimization (MOPSO) algorithm is applied for parameter identification. The optimized model parameters are immediately adopted to predict the excavation-induced pile deformation in the continuous construction stages. To achieve efficient parameter optimization and real-time prediction of system behavior, the back propagation neural network (BPNN) is established to substitute the finite element model, which is further implemented together with MOPSO for automatic operation. The proposed approach is applied in the Taihu tunnel excavation project, where the effectiveness of the method is demonstrated via the comparisons with the site monitoring data. The method is reliable with a prediction accuracy of more than 90%. Moreover, different optimization algorithms, including non-dominated sorting genetic algorithm (NSGA-II), Pareto Envelope-based Selection Algorithm II (PESA-II) and MOPSO, are compared, and their influences on the prediction accuracy at different excavation stages are studied. The results show that MOPSO has the best performance for high dimensional optimization task.

Numerical back analysis method of three-dimensional in situ stress fields considering complex surface topography and variable collinearity

Numerical back analysis based on incomplete in situ stress measurements is widely used to estimate the three-dimensional (3D) in situ stress fields of large deep underground caverns. However, the variable collinearity caused by complex geological environments has rarely been considered. This paper proposes a numerical back analysis method for 3D in situ stress fields based on stepwise regression (BSSR). In BSSR, the collinearity between independent variables is associated with the surface topography, and insignificant variables caused by variable collinearity are eliminated through stepwise regression. BSSR was applied to the underground powerhouse of a hydropower station for validation, and the results showed that it can reliably estimate the 3D in situ stress field despite the complex geological environments. Multidimensional mathematical models of in situ stress fields were established with the improved predictive ability and clear physical meaning, which can quantify the contributions of six geological actions and three stress sources. The findings of this study can help with understanding the formation mechanism of in situ stress fields for large, deep underground caverns in complex geological environments and provide a useful reference for optimizing excavation schemes and support designs.

Probabilistic back analysis of rainfall-induced slope failure considering slope survival records from past rainfall events

Many rainfall-induced landslides occur on slopes that have limited, or even no, site investigation data before failure. Probabilistic back analysis of slope failure provides an effective tool to back analyze possible pre-failure soil parameters, thus gaining insights into the mechanism of slope failure. For a slope failure induced by rainfall, the actual factor of safety (FS) is time-variant when time-variant rainfall and infiltration are explicitly modeled. This study proposes a novel probabilistic back analysis method that models explicitly the rainfall triggering mechanism for a rainfall-induced slope failure. Unlike existing methods that are based on a constant FS = 1, the proposed method utilizes FS inequality information for probabilistic back analysis, including slope failure record with FS < 1 and slope survival records from past rainfall events with FS > 1. The proposed method is suitable for high-dimensional problems when soil hydraulic properties are also back analyzed. The proposed method converges to the conventional methods with FS = 1 when the time span of slope failure is narrowed down to a few hours and slope survival records are ignored. Incorporating both slope failure and survival records effectively reduces uncertainties in soil strength and hydraulic parameters.

Back analysis of geomechanical parameters for rock mass under complex geological conditions using a novel algorithm

Geomechanical properties of rock mass are of vital importance for design optimization and stability evaluation in geotechnical engineering. Traditionally, they are estimated by laboratory experiments, in situ tests, and empirical classification, which ignore the scale effect, and may be not representative, leading to uncertainties of parameters. Back analysis method based on the monitoring displacement data provides the state-of-the-art tool for the estimation of geomechanical parameters. This paper proposed a novel back analysis method, which combined the optimized particle swarm optimization (OPSO) algorithm with the support vector machine (SVM) algorithm. And then, we cooperate the method with finite element method (FEM) forming the comprehensive method named OPSO-SVM-ABAQUS. This method has developed the standard PSO algorithm from two aspects for more powerful optimization capability and higher optimization precision. The OPSO algorithm is then used for the estimation of the SVM hyperparameters and the optimal values of geomechanical parameters. Furthermore, the performance of the proposed algorithm is compared with other six algorithms. Finally, the geomechanical parameters of the interlayered soft and hard rocks are calculated accurately, with the average error between the back-analyzed and assumed actual parameters to be 0.91 %. And the parametric investigations are implemented with the trained model. It dominates that the proposed algorithm is feasible for reasonable parameters in geotechnical engineering under complex geological conditions.

An Experimental Study of Industrial Site and Shaft Pillar Mining at Jinggezhuang Coal Mine

Engineering site and shaft pillars are excavated to prolong the life of collieries and excavate more underground coal. The Jinggezhuang colliery (‘JGZ’) is a resource-exhausted coal mine in eastern China. It was determined that the industrial site and shaft pillar of JGZ would be extracted in 2008. This study excavated an experimental panel to examine the effect of pillar excavation on surface buildings in complicated geological conditions. A new pillar design was proposed based on surface monitoring to increase the recovery ratio. To maintain the safety of the shaft and engineering facilities, panel 0091 was mined and surface deformation was monitored during the experiment. The deformation characteristics and parameters were obtained using a back analysis method. A new pillar was designed using the parameters measured from panel 0091. The design maintained the safety of the shaft but relaxed the restriction of the influence of constructions at the engineering site. The prediction results of the surface subsidence and the deformation of the main building were analyzed. The maximum subsidence of the surface was 7419 mm, but the surface subsidence of the shafts was less than 10 mm. The shafts were weakly influenced by the pillar excavation. The prediction results can be used as basic information for the monitoring and maintenance of buildings in the future. Using the new pillar design, 2.54 million tons of coal resources were mined. This study provides an engineering example and a reference for shaft pillar excavation in the future.

Design of Tunnel Initial Support in Silty Clay Stratum Based on the Convergence-Confinement Method

The stress release ratio of the surrounding rock in tunnel excavation is one of the most important indicators that affect the stress distribution and displacement of the surrounding rock. To determine the variation law of the stress release ratio of the surrounding rock during excavation in silty clay stratum, the stress release law is determined based on the convergence–confinement method (CCM) and field test data. The stress release law of the surrounding rock under support is determined based on the displacement back analysis method. The permitted displacement safety factor of silty clay under different subgrade conditions and the optimal supporting time of the initial supporting structure are determined by comparing the stress release ratio with surrounding rock displacement. The results indicated that the stress release ratio of surrounding rock in the silty clay stratum is approximately 78–90% when the coordinate displacement of the supporting structure and surrounding rock is stable under the current excavation and support conditions. For the surrounding rock of subgrade V in the silty clay stratum, the safety factor of the permitted displacement in the tunnel vault is approximately 2.91, and the initial support should be carried out within 1 m behind the face advancing. For the surrounding rock of subgrade VI1, the safety factor of the permitted displacement is 1.40, and the initial support must be carried out 1 m ahead of the tunnel face. For the surrounding rock of grade VI2, the initial support must be carried out 4 m ahead of the tunnel face.

Back-Analysis of Slope GNSS Displacements Using Geographically Weighted Regression and Least Squares Algorithms

Numerical simulation is a powerful technique for slope stability assessment and landslide hazard investigation. However, the physicomechanical parameters of the simulation results are susceptible to uncertainty. Displacement back-analysis is considered an effective method for the prediction of the geomechanical parameters of numerical models; therefore, it can be used to deal with the parameter uncertainty problem. In this study, to improve the interpretability of the back-analysis model, an analytical function relationship between slope displacements and physicomechanical parameters was established using geographically weighted regression. By combining the least-squares and linear-algebra algorithms, a displacement back-analysis method based on geographically weighted regression (DBA-GWR) was developed; in particular, the multi-objective displacement back-analysis was represented as an analytical problem. The developed method was subsequently used for a slope of the Guiwu Expressway in Guangxi, China. Simulation experiments and GNSS real-data experiments demonstrated that the GWR could achieve high-precision deformation modelling in the spatial domain with model-fitting precision in the order of mm. Compared with state-of-the-art methods, the precision of the simulated displacement with the proposed method was significantly improved, and equivalent physicomechanical parameters with higher accuracy were obtained. Based on the corrected numerical model, the most severely deformed profiles were forward-analysed, and the simulated deformation and distribution patterns were found to be in good agreement with the field investigation results. This approach is significant for the determination of geomechanical parameters and the accurate assessment of slope safety using monitoring data.