Prediction Of Liquefaction Research Articles

Implementation of machine learning classification models considering the optimum data ratio in predicting soil liquefaction susceptibility

Abstract Soil Liquefaction has a disastrous impact on structures and underground infrastructure. Therefore, an appropriate liquefaction vulnerability assessment strategy can help reduce the detrimental consequences of this hazard. In recent decades, machine learning has been studied more frequently to solve geotechnical issues, such as determining liquefaction susceptibility. Intending to improve the model’s learning ability to identify liquefaction vulnerability and to find the optimum training and testing data ratio, this research attempts to develop a machine learning model for liquefaction prediction utilizing relatively more varied data in different data ratios. In this study, liquefaction prediction models were developed using four supervised learning-based algorithms: Random Forest (RF), Naïve Bayes Classifier (NBC), Decision Tree (DT), and K-Nearest Neighbor (k-NN). Seven parameters were utilized to train the model using historical data on liquefaction. The model’s performance in predicting liquefaction was compared with various training and testing data ratios and validated using 5-fold cross-validation. The capability of the model was assessed using performance metrics. The results show that the RF model has the highest accuracy in predicting liquefaction among all the algorithms used. RF achieved an overall accuracy of 90.28%, followed by the k-NN (86.11%) and the DT (81.94%) on a training and testing data ratio of 80:20. The NBC algorithm obtained the highest accuracy of 78.44% on the 75:25 data ratio. In general, the machine learning approach is capable of predicting liquefaction susceptibility.

Soil Categorization and Liquefaction Prediction Using Deep Learning and Ensemble Learning Algorithms

Soil Categorization and Liquefaction Prediction Using Deep Learning and Ensemble Learning Algorithms

Evaluation of liquefaction potential in central Taiwan using random forest method

Liquefaction is a significant geotechnical hazard in seismically active regions like Taiwan, threatening infrastructure and public safety. Accurate prediction models are essential for assessing soil susceptibility to liquefaction during seismic events. This study evaluates liquefaction potential in central Taiwan using the random forest (RF) method. The RF models were developed with a dataset of 540 soil and seismic parameter sets, including depth, effective and total overburden stresses, SPT-N values, fine soil content, earthquake magnitude, peak ground acceleration, and historical liquefaction occurrences. Rigorous validation techniques, such as cross-validation and comparisons with observed liquefaction events, confirm the RF model’s effectiveness, achieving an accuracy of 98.89%. The model also quantifies predictor importance, revealing that the SPT-N value is the most critical soil factor, while peak ground acceleration is the key seismic factor for liquefaction prediction. Notably, the RF model outperforms simplified procedures in accuracy, even with fewer input factors. Our case studies show that an accuracy of over 95% can still be achieved, highlighting the RF model’s superior performance compared to conventional methods, which struggle to reach similar levels.

Developing Liquefaction Prediction Equations for Gravels Based on DPT Blow Count and Shear Wave Velocity at Field Case History Sites

Gravelly soils have liquefied at multiple sites in at least 20 earthquakes over the past 130 years. These gravels typically contain more than 25% sand which lowers the permeability and makes them susceptible to liquefaction. In other cases, a low permeability layer at the surface impedes drainage and causes liquefaction. Typical SPT- or CPT-based liquefaction correlations can be affected by large gravel particles and lead to erroneous results. To deal with these problems, we have developed liquefaction triggering curves for gravelly soils based on (1) shear wave velocity (Vs) and (2) a 74-mm diameter dynamic cone penetrometer (DPT), that are less affected by gravel particles. These correlations are based on case histories where gravel did and did not liquefy in past earthquakes. The VS-based liquefaction triggering curves for gravels shift to the right relative to similar curves based on sands. Good agreement has been obtained with liquefaction resistance from the DPT and the CPT when these methods could all penetrate the gravel in Wellington New Zealand. Correlations are developed between Vs and blow count from the DPT.

Artificial Neural Networks and Ensemble Learning for Enhanced Liquefaction Prediction in Smart Cities

This paper examines how smart cities can address land subsidence and liquefaction in the context of rapid urbanization in Japan. Since the 1960s, liquefaction has been an important topic in geotechnical engineering, and extensive efforts have been made to evaluate soil resistance to liquefaction. Currently, there is a lack of machine learning applications in smart cities that specifically target geological hazards. This study aims to develop a high-performance prediction model for estimating the depth of the bearing layer, thereby improving the accuracy of geotechnical investigations. The model was developed using actual survey data from 433 points in Setagaya-ku, Tokyo, by applying two machine learning techniques: artificial neural networks (ANNs) and bagging. The results indicate that machine learning offers significant advantages in predicting the depth of the bearing layer. Furthermore, the prediction performance of ensemble learning improved by about 20% compared to ANNs. Both interdisciplinary approaches contribute to risk prediction and mitigation, thereby promoting sustainable urban development and underscoring the potential of future smart cities.

Enhancing soil liquefaction risk assessment with metaheuristics and hybrid learning techniques

ABSTRACT Soil liquefaction has garnered significant research attention over several decades due to its profound impact on infrastructure systems in earthquake-prone regions. Given the limitations inherent in the assumptions and approximations of conventional methods, machine learning has emerged as a robust and efficient approach for predicting soil liquefaction potential. This study aims to develop a synergistic JS-CNN-XGB model, unifying the Jellyfish Search (JS) optimizer with a hybrid deep/machine learning model. This amalgamation leverages the feature extraction capabilities of the Convolutional Neural Network (CNN) alongside the classification prowess of the eXtreme Gradient Boosting (XGB) algorithm. The proposed model seamlessly integrates into a user-friendly prediction system, streamlining the liquefaction prediction process. Validation is achieved through case studies involving historical earthquake recordings with diverse classification ratios and varying input attribute quantities. Compared to existing studies, our proposed system showcases a notable 6.2% enhancement in overall liquefaction assessment accuracy, demonstrating improved detection capabilities in imbalanced and balanced datasets. In conclusion, this study underscores an automated system that presents a robust solution to the challenges posed by liquefaction potential assessment in geotechnical engineering.

Simulating liquefaction-induced resuspension flux in a sediment transport model

Wave-induced liquefaction is a geological hazard under the action of cyclic wave load on seabed. Liquefaction influences the suspended sediment concentration (SSC), which is essential for sediment dynamics and marine water quality. Till now, the identification of liquefaction state and the effect of liquefaction on SSC have not been sufficiently accounted for in the sediment model. In this study, we introduced a method for simulating the liquefaction-induced resuspension flux into an ocean model. We then simulated a storm north of the Yellow River Delta, China, and validated the results using observational data, including significant wave heights, water levels, excess pore water pressures, and SSCs. The liquefaction areas were mainly distributed in coastal zones with water depths less than 12 m, and the simulated maximum potential soil liquefaction depth was 1.39 m. The liquefaction-induced SSC was separated from the total SSC of both liquefaction- and shear-induced SSCs by the model, yielding a maximum liquefaction-induced SSC of 1.07 kg·m−3. The simulated maximum proportion of liquefaction-induced SSC was 26.2% in regions with water depths of 6–12 m, with a maximum significant wave height of 3.4 m along the 12 m depth contour. The erosion zone at water depths of 8–12 m was reproduced by the model. Within 52.5 h of the storm, the maximum erosion thickness along the 10 m depth contour was enhanced by 33.9%. The model is applicable in the prediction of liquefaction, and provides a new method to simulate the SSC and seabed erosion influenced by liquefaction. Model results show that liquefaction has significant effects on SSC and seabed erosion in the coastal area with depth of 6–12 m. The validity of this method is confined to certain conditions, including a fully saturated seabed exhibiting homogeneity and isotropic properties, small liquefaction depth, residual liquefaction dominating the development of pore pressures, no influence by structures, and the sediment composed of silt and mud that experiences frequent wave-induced liquefaction.

Accurate and generalizable soil liquefaction prediction model based on the CatBoost algorithm

Accurate and generalizable soil liquefaction prediction model based on the CatBoost algorithm

Analisa Prediksi Potensi Likuifaksi di Kota Rengat Kabupaten Indragiri Hulu

The number of incidents occurring on the island of Sumatra, especially West Sumatera and Jambi provinces that affect the vibration of the earthquake in Rengat City on a large scale can lead to the movement of land such as landslides and stagnant ground water sources can lead to liquefaction. The purpose of this research is to know how the arrangement of soil type at research location Rengat City and to know the location reviewed in Rengat City, and its surrounding experience potential liquefaction based on some existing penetration test data. This research was conducted to predict liquefaction potential that occurred in Rengat City of Indragiri Hulu Regency with secondary data of penetration test of konus with Seed and Idriss method (1971). The calculation of potential liquefaction prediction is done by determining the number of soil layers, estimating the weight of the soil volume, determining the soil overburden, determining the effective soil stress, determining the corrected confront resistance, calculating Cyclic Strees Ratio (CSR), calculating Cyclic Resistance Ratio (CRR). After calculating the value of Magnitude Scalling Factor (MSF) and calculating the value of safe factor by comparing the value of Cyclic Strees Ratio (CSR) with Cyclic Resistance Ratio (CRR) if the calculation <1 factor is safe then potentially liquefaction occurs. Based on the results of the analysis of the 7 locations with a total of 2 cone penetration test conus analyzed in this study in Rengat City, obtained the type of soil structure is not uniform and based on the results of analysis, from the 7 research sites with tottal 21 point penetration test konus reviewed at magnitude 5 to 10 in Rengat City Indragiri Hulu Regency. In the type of sand soil in the dominant study area can not be achieved liquefaction, this is reinforced from the calculation (SF)> 1 safe factor (SF: 1.0).

A comparative analysis of ensemble learning algorithms with hyperparameter optimization for soil liquefaction prediction

A comparative analysis of ensemble learning algorithms with hyperparameter optimization for soil liquefaction prediction

Prediction of liquefaction of gravelly soils based on a cost-sensitive Bayesian network combined with rough set weighting

Existing liquefaction prediction approaches for gravelly soils rely on in situ test data, which neglects the in-suit test cost sensitivity. Furthermore, these approaches do not consider the effect of the weights of the factors on the model performance. Therefore, in this paper, a cost-sensitive Bayesian network with rough set theory (CSW-BN) and model hierarchies and factor weighting are combined to address the two aforementioned challenges in research on gravelly soil liquefaction prediction. A two-layer model is constructed using the proposed CSW-BN approach. The first layer of the model reduces the data costs by utilizing conventional soil parameters, without necessitating specific in situ tests such as shear wave velocity tests. In the second layer of the model, variable weights calculated using rough set theory are incorporated into the parameter learning to enhance the generalization of the model. The results show that the learning and prediction accuracies of the new approach are 0.968 and 0.95, respectively. Compared with existing models, the proposed method not only reduces the number of required in situ tests but also improves the prediction accuracy and interpretability. Furthermore, the effectiveness of the CSW-BN model is evaluated using new cases for the 2008 Wenchuan earthquake, with a prediction accuracy of 0.89. Finally, a software tool is developed using Visual Basic for Applications programming to facilitate its application in engineering practice.

The influence of anthropogenic regulation and evaporite dissolution on earthquake-triggered ground failure

Remote sensing observations of Searles Lake following the 2019 moment magnitude 7.1 Ridgecrest, California, earthquake reveal an area where surface ejecta is arranged in a repeating hexagonal pattern that is collocated with a solution-mining operation. By analyzing geologic and geotechnical data, here we show that the hexagonal surface ejecta is likely not a result of liquefaction. Instead, we propose dissolution cavity collapse (DCC) as an alternative driving mechanism. We support this theory with pre-event Interferometric Synthetic Aperture Radar data, which reveals differential subsidence patterns and the creation of subsurface void space. We also find that DCC is likely triggered at a lower shaking threshold than classical liquefaction. This and other unknown mechanisms can masquerade as liquefaction, introducing bias into liquefaction prediction models that rely on liquefaction inventories. This paper also highlights the opportunities and drawbacks of using remote sensing data to disentangle the complex factors that influence earthquake-triggered ground failure.

Probabilistic evaluation of earthquake-induced liquefaction using Bayesian network based on a side-by-side SPT–CPT database

In situ tests such as standard penetration test (SPT) and cone penetration test (CPT) are often conducted to evaluate the probability of earthquake-induced liquefaction. However, the models adopted have seldom attempted to utilize SPT and CPT data simultaneously. In this study, a side-by-side SPT–CPT database at historical earthquake sites is established; then, a Bayesian network model is constructed to predict the probability of soil liquefaction based on this database, with which the SPT and CPT data are utilized simultaneously. Next, comparative studies are undertaken to illustrate the superiority of the Bayesian network-based probabilistic soil liquefaction model developed over other models, in terms of six SPT- and CPT-based conventional liquefaction models in the literature and two Bayesian network-based models. It should be noted that the liquefaction sites with two in situ tests are scarce and side-by-side SPT–CPT data can be incomplete, which leads to challenges in applying the Bayesian network model developed. To address this problem, correlations between SPT and CPT data are analyzed, and these correlations are further included in the Bayesian network model; as a result, a modified Bayesian network model is reached. Finally, the influence of the proportion of missing data in the incomplete SPT–CPT data on the liquefaction prediction accuracy is discussed.

Accelerated modeling and design of a mixed refrigerant cryogenic process using a data-driven approach

Cryogenic processes with mixed refrigerants are prevalent in energy-intensive chemical industries, enhancing energy efficiency while reducing costs and unit size. However, the curse of dimensionality and process design constraints pose significant hurdles for effective screening and optimization. To tackle this, we developed a neural network model for natural gas liquefaction prediction. Trained on an extensive Aspen HYSYS database, our ML model accurately simulates LNG processes, with an impressive R2 test value of 99.63, operating almost ten million times faster than HYSYS. It effectively addresses vital process design constraints, including liquid slugging and temperature cross, crucial for optimization. By integrating the ML model with genetic and Nelder–Mead algorithms, we achieve an 8.9% reduction in total exergy, outperforming Aspen HYSYS within the same time frame. Our study underscores ML’s significance in modeling energy-intensive chemical processes, providing insights into the exergy profile and enabling feature importance analysis.

Geospatial Liquefaction Modeling of the 2023 Türkiye Earthquake Sequence by an Ensemble of Global, Continental, Regional, and Event-Specific Models

Abstract A global geospatial liquefaction model (GGLM-2017) was previously developed (Zhu et al., 2017) using logistic regression (LR) and is currently used by the U.S. Geological Survey as the preferred liquefaction model to map liquefaction probability immediately after the occurrence of earthquake events. This research proposes an ensemble modeling approach to improve the performance of the GGLM-2017 for geospatial liquefaction modeling of the 2023 Türkiye earthquakes using an updated inventory of liquefaction occurrence locations in Europe (the OpenLIQ database, which includes prior events occurring in Türkiye) and a new inventory from the 2023 Türkiye earthquakes (gathered from multiple sources). Using the same geospatial proxies for soil saturation, soil density, and earthquake loading, and the same non-liquefaction sampling strategy used to develop GGLM-2017, the proposed ensemble method is validated on the data of the 2023 Türkiye earthquakes by integrating four models, including global (GGLM-2017), continental (LR model trained on eight European events), regional (LR model trained on three historical events in Türkiye), and event-specific (LR model trained on partially available data from the 2023 Türkiye earthquakes) models. The inventory from the 2023 Türkiye earthquakes is split into two batches, in which the first batch (163 liquefaction occurrences) resulted from the preliminary reconnaissance and is used for training the event-specific model, and the second batch (284 liquefaction occurrences) resulted from a more complete reconnaissance (which was made available later) and is used for validating all models. The rationale for using the first batch for training the event-specific model is to exploit the information as they become available to optimize the performance of global model in liquefaction prediction. The final ensemble probability assignment is done by averaging the probabilities derived by the four individual models, and a 50% threshold is used for classification accuracy evaluations. Comparative analysis of the ensemble model’s performance with the GGLM-2017 showed improved predictive accuracy, resulting in higher liquefaction detection for the specific event under study (the 2023 Türkiye earthquakes). The ensemble model also provides an estimate of model uncertainty.

Challenges to Innovation of Liquefaction Prediction Technology

Challenges to Innovation of Liquefaction Prediction Technology

Semi-Supervised Learning Method for the Augmentation of an Incomplete Image-Based Inventory of Earthquake-Induced Soil Liquefaction Surface Effects

Soil liquefaction often occurs as a secondary hazard during earthquakes and can lead to significant structural and infrastructure damage. Liquefaction is most often documented through field reconnaissance and recorded as point locations. Complete liquefaction inventories across the impacted area are rare but valuable for developing empirical liquefaction prediction models. Remote sensing analysis can be used to rapidly produce the full spatial extent of liquefaction ejecta after an event to inform and supplement field investigations. Visually labeling liquefaction ejecta from remotely sensed imagery is time-consuming and prone to human error and inconsistency. This study uses a partially labeled liquefaction inventory created from visual annotations by experts and proposes a pixel-based approach to detecting unlabeled liquefaction using advanced machine learning and image processing techniques, and to generating an augmented inventory of liquefaction ejecta with high spatial completeness. The proposed methodology is applied to aerial imagery taken from the 2011 Christchurch earthquake and considers the available partial liquefaction labels as high-certainty liquefaction features. This study consists of two specific comparative analyses. (1) To tackle the limited availability of labeled data and their spatial incompleteness, a semi-supervised self-training classification via Linear Discriminant Analysis is presented, and the performance of the semi-supervised learning approach is compared with supervised learning classification. (2) A post-event aerial image with RGB (red-green-blue) channels is used to extract color transformation bands, statistical indices, texture components, and dimensionality reduction outputs, and performances of the classification model with different combinations of selected features from these four groups are compared. Building footprints are also used as the only non-imagery geospatial information to improve classification accuracy by masking out building roofs from the classification process. To prepare the multi-class labeled data, regions of interest (ROIs) were drawn to collect samples of seven land cover and land use classes. The labeled samples of liquefaction were also clustered into two groups (dark and light) using the Fuzzy C-Means clustering algorithm to split the liquefaction pixels into two classes. A comparison of the generated maps with fully and manually labeled liquefaction data showed that the proposed semi-supervised method performs best when selected high-ranked features of the two groups of statistical indices (gradient weight and sum of the band squares) and dimensionality reduction outputs (first and second principal components) are used. It also outperforms supervised learning and can better augment the liquefaction labels across the image in terms of spatial completeness.

Prediction of soil liquefaction for railway embankment resting on fine soil deposits using enhanced machine learning techniques

Prediction of soil liquefaction for railway embankment resting on fine soil deposits using enhanced machine learning techniques

Calibration of Pore Pressure Model for the Liquefaction Prediction of Soils Based on Behavior of Clean Sand, Magnitude Scaling Factor, and Initial Stress Level

Calibration of Pore Pressure Model for the Liquefaction Prediction of Soils Based on Behavior of Clean Sand, Magnitude Scaling Factor, and Initial Stress Level

Numerical Simulations of Soil Liquefaction using PLAXIS 2D

One of the most serious and complex phenomena that can occur during earthquakes is soil liquefaction. Since Taiwan is located in the seismic zone of the Pacific Rim, where earthquakes and strong earthquakes occur frequently, the evaluation and prediction of liquefaction have attracted the attention of many authorities, researchers, and engineers in Taiwan. Therefore, in this paper, PLAXIS 2D software was used in conjunction with two model conditions, with and without deep excavation project, and two types of constitutive models, Hardening Soil model with small-strain stiffness (HSS) and UBC3D-PLM, to simulate a series of numerical analyses to predict liquefaction potential. The soil profile of Sanchong District, one of the metropolitan areas in Taiwan considered to be particularly susceptible to soil liquefaction, was selected as the soil profile. The results show that the UBC3D-PLM model has a relatively higher liquefaction potential than the Hardening Soil model with small-strain stiffness (HSS) for the model condition without the deep excavation project. However, the dynamic calculation using UBC3D-PLM model cannot be fully performed for the model condition with the deep excavation project.