Liquefaction Risk Assessment Research Articles

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.

Probabilistic liquefaction risk assessment using the ground settlement for the Japan-wide area

Probabilistic liquefaction risk assessment using the ground settlement for the Japan-wide area

A hyper parameterized artificial neural network approach for prediction of the factor of safety against liquefaction

Soil liquefaction during earthquakes is a complex geotechnical engineering problem. Although various analytical approaches exist for predicting liquefaction risk, their limitations have led researchers to explore using artificial intelligence and machine learning methods. Machine learning has the potential to significantly improve the ability to predict soil liquefaction and mitigate the associated risks. This study proposes a hyper parameterized artificial neural network architecture using random search, grid search, and Bayesian optimization algorithms to predict the factor of safety against liquefaction. The performances of hyper parameterized machine learning algorithms, including artificial neural networks (ANN), decision trees (DT), random forest (RF), and support vector regression (SVR), were compared. Statistical tests show that the proposed ANN outperformed the others with a determination coefficient of 0.99 at a 95% significance level. Hyperparameter optimization significantly improved learning performance with up to a 48% reduction in RMSE scores. The proposed method was compared with previous studies, and performance results confirmed its effectiveness and generalization ability. In conclusion, this study highlights the potential of machine learning algorithms for predicting soil liquefaction and emphasizes the importance of hyperparameter optimization for improving model performance. The findings of this study have practical implications for improving liquefaction risk assessment and mitigating the associated hazards.

Three elements of liquefaction risk of liquefiable solid bulk cargoes during sea transport: Critical review

The transport of liquefiable solid bulk cargo is one of the main causes of maritime safety accidents over the past decade. This article innovatively summarized the status quo of research on the liquefaction risk of liquefiable solid bulk cargo during sea transport from three aspects, namely, the properties, moisture content and external induced load of cargoes during their sea transport. The influence of the main physical parameters, the application of modern testing techniques (e.g., transparent soil and PIV) in related fields, and various TML testing methods in IMSBC code were also introduced. Result highlighted external induced load as an important factor affecting the cargo liquefaction during sea transport. The latest trends on the adoption of the liquefaction risk assessment method based on external load were also systematically introduced, and the importance of quantitative risk assessment in practical application was emphasized. This article also analyzed the main challenges encountered in managing liquefaction risk and revealed quantitative risk assessment and liquefaction mechanism as the main trends in related research. Four promising directions for future research were also put forward, which can provide valuable ideas for researchers and managers to further study and control the liquefaction risk of liquefiable solid bulk cargo.

Reliability Analysis for Liquefaction Risk Assessment for the City of Patna, India using Hybrid Computational Modeling

ABSTRACT In the present study, the first-order reliability method (FORM) is applied to evaluate the failure of soil deposits during seismic excitation for the city of Patna, India. Patna is emerging as one of the metro cities and the rapid infrastructure development in the city with high pace construction of road and metro services along with several smart city projects have led to immense growth in civil engineering structures. Therefore, liquefaction assessment of Patna is an important subject due to the geographical and seismic location of the city. A detailed comparative study has been performed between first-order second moment (FOSM) and advanced first-order second-moment (AFOSM) reliability methods to determine the most suitable method to evaluate the potential risk of liquefaction for Patna city. Reliability index (β) values obtained from AFOSM analysis are in true accordance with the deterministic approach and therefore can be considered as an appropriate tool for reliability analysis for this city. The analysis establishes that the city of Patna exhibits high possibilities of liquefaction failure during high-intensity earthquakes i.e. Mw = 6.5. Also, a concept of a predictive computational model developed by the hybridization of ANN and GWO algorithms to determine β value using geotechnical and seismic parameters has been proposed. The high precision and error-free performance of the ANN-GWO model provides a powerful computational tool to assist the prediction of β. The results of the study could be used to comprehend the potential risk against liquefaction and provide a novel and insightful concept of risk assessment for safe and economic construction practices.

Seismic Liquefaction of Saturated Calcareous Sands: Dynamic Centrifuge Test and Numerical Simulation

Both a dynamic centrifuge test and dynamic finite element analysis were carried out to assess the seismic liquefaction risk of a saturated-calcareous-sand site in a port project in Timor-Leste. Taking the in situ calcareous sands as the model material, two groups of horizontal free-field model tests for medium dense and dense saturated calcareous sands were completed based on the two-stage scaling law theory. Three natural earthquake records with varied peak accelerations were adopted as input motions. The experimental results indicate that the shallower the depth, the lower the relative density, the longer the seismic duration, the larger the peak acceleration, and the more susceptible the saturated site to liquefaction. The sands on site at a depth of five to ten meters is highly risky for liquefaction with the excess pore pressure ratio reaching up to about 1.0 under the seismic peak acceleration of 0.3 g. The risk of liquefaction for site sands is rather small under the seismic peak acceleration of 0.1 g. The study reveals the characteristics of the pore pressure development in sites of varied relative densities under different seismic loadings, which provides a scientific basis for the liquefaction risk assessment of the engineering site.

Seismic Liquefaction Risk Assessment of Critical Facilities in Kathmandu Valley, Nepal

Kathmandu Valley lies in an active tectonic zone, meaning that earthquakes are common in the region. The most recent was the Gorkha Nepal earthquake, measuring 7.8 Mw. Past earthquakes caused soil liquefaction in the valley with severe damages and destruction of existing critical infrastructures. As for such infrastructures, the road network, health facilities, schools and airports are considered. This paper presents a liquefaction susceptibility map. This map was obtained by computing the liquefaction potential index (LPI) for several boreholes with SPT measurements and clustering the areas with similar values of LPI. Moreover, the locations of existing critical infrastructures were reported on this risk map. Therefore, we noted that 42% of the road network and 16% of the airport area are in zones of very high liquefaction susceptibility, while 60%, 54%, and 64% of health facilities, schools and colleges are in very high liquefaction zones, respectively. This indicates that most of the critical facilities in the valley are at serious risk of liquefaction during a major earthquake and therefore should be retrofitted for their proper functioning during such disasters.

A step forward towards a comprehensive framework for assessing liquefaction land damage vulnerability: Exploration from historical data

The unprecedented liquefaction-related land damage during earthquakes has highlighted the need to develop a model that better interprets the liquefaction land damage vulnerability (LLDV) when determining whether liquefaction is likely to cause damage at the ground’s surface. This paper presents the development of a novel comprehensive framework based on select case history records of cone penetration tests using a Bayesian belief network (BBN) methodology to assess seismic soil liquefaction and liquefaction land damage potentials in one model. The BBN-based LLDV model is developed by integrating multi-related factors of seismic soil liquefaction and its induced hazards using a machine learning (ML) algorithm-K2 and domain knowledge (DK) data fusion methodology. Compared with the C4.5 decision tree-J48 model, naive Bayesian (NB) classifier, and BBN-K2 ML prediction methods in terms of overall accuracy and the Cohen’s kappa coefficient, the proposed BBN K2 and DK model has a better performance and provides a substitutive novel LLDV framework for characterizing the vulnerability of land to liquefaction-induced damage. The proposed model not only predicts quantitatively the seismic soil liquefaction potential and its ground damage potential probability but can also identify the main reasons and fault-finding state combinations, and the results are likely to assist in decisions on seismic risk mitigation measures for sustainable development. The proposed model is simple to perform in practice and provides a step toward a more sophisticated liquefaction risk assessment modeling. This study also interprets the BBN model sensitivity analysis and most probable explanation of seismic soil liquefied sites based on an engineering point of view.

Determination of Initial-Shear-Stress Impact on Ramsar-Sand Liquefaction Susceptibility through Monotonic Triaxial Testing

Liquefaction risk assessment is critical for the safety and economics of structures. As the soil strata of Ramsar area in north Iran is mostly composed of poorly graded clean sand and the ground water table is found at shallow depths, it is highly susceptible to liquefaction. In this study, a series of isotropic and anisotropic consolidated undrained triaxial tests were performed on reconstituted specimens of Ramsar sand to identify the liquefaction potential of the area. The specimens are consolidated isotropically to simulate the level ground condition, and anisotropically to simulate the soil condition on a slope and/or under a structure. The various states of soil behavior are studied by preparing specimens at different initial relative densities and applying different levels of effective stress. The critical state soil mechanics approach for identifying the liquefaction susceptibility is adopted and the observed phenomena are further explained in relation to the micro-mechanical behavior. As only four among the 27 conducted tests did not exhibit liquefactive behavior, Ramsar sand can be qualified as strongly susceptible to liquefaction. Furthermore, it is observed that the pore pressure ratio is a good indication of the liquefaction susceptibility.

A simplified vulnerability model for the extensive liquefaction risk assessment of buildings

The quantitative assessment of natural risks offers a rational strategy to protect communities, undertake cost effective mitigation and plan the organic and sustainable development of urban systems. For cascade events such as earthquake-induced liquefaction, assessment implies to characterize and reconstruct the areal distribution of seismic hazard, subsoil susceptibility, physical vulnerability, economic and social relevance of structures and to combine all factors in a unitary predictive model. Considering that aleatory variability and epistemic uncertainty affect the characteristic variables and their mutual correlation, it is also necessary to quantify their influence on the prediction. Within this framework, a vulnerability model is proposed to comprehensively assess the physical damage of buildings in an urban system. A chain method is formulated combining calculation schemes recently introduced in the literature with ad hoc numerical analyses. The effectiveness of the method is tested comparing prediction with the effects observed in the city of Christchurch during the 22nd February 2011 earthquake. The unprecedented documentation available after this earthquake enables to validate different components of the model and disclose the importance of possible disregarded factors. A geostatistical methodology is proposed throughout the paper to process data, quantify and govern the different uncertainty factors.

A next-generation open-source tool for earthquake loss estimation

The present research has been benefited from funding of NORSAR and the Univ. Alicante through research contracts (NORSAR1-14A, NORSAR1-08I), the funding of the Ministerio de Economia, Industria y Competitividad (CGL2016-77688-R) and the Generalitat Valenciana (BEST/2012/173 and AICO/2016/098). The development and implementation of the liquefaction risk assessment methodology is done under the LIQUEFACT project funded by the European Union’s Horizon 2020 research and innovation programme under grant agreement (No. 700748).

LIQUEFACTION RISK ASSESSMENT BY THE USE OF GEOPHYSICAL TECHNIQUES: THE TEST AREA OF NAFPLION CITY, GREECE

An efficient and cost effective site characterization, with regard to the seismic hazard and liquefaction risk assessment, was accomplished with the aid of geophysics in the area, where the Nafplion city of Greece is expanding. The methodology adopted includes the recognition of the possible earthquake sources of the wider region, their modelling, in order to stochastically simulate the strong ground motion at the investigation area, and finally the calculation of the liquefaction risk. The investigation area was suspected of high liquefaction potential since the foundation ground consists of loose sandy silt with very shallow aquifer. The geophysical techniques considerably contributed to the detection and characterization of possible local seismic faults with the implementation of gravity and seismic methods. Special emphasis was given to the seismic depth migration and particularly to the construction of valid velocity models, in order to precisely calculate the dips of the possible faults. Additionally the geophysical techniques provided the near surface velocity structure for the calculation of the amplification of the seismic motion up to the surface, also required for the final estimation of the liquefaction risk. The seismic methods (seismic reflection, seismic refraction, seismic modelling, MASW, multichannel analysis of microtremors and crosshole investigations), if combined with geo-technical borehole testing, enhance their reliability and cover large areas in a cost-effective way in comparison with the standard borehole tests. In Nafplion area, evidence was found for a low factor of safety against liquefaction at specific sites within the study area. The results show that liquefaction probability can reach 80% at some sites depending on selected earthquake scenario, mainly at depths between 5 and 10 meters. This should be considered as highly important information for making risk-based design decision in this region.

Assessment of Stone Columns as a Mitigation Technique of Liquefaction-Induced Effects during Italian Earthquakes (May 2012)

Soil liquefaction has been observed worldwide during recent major earthquakes with induced effects responsible for much of the damage, disruption of function, and considerable replacement expenses for structures. The phenomenon has not been documented in recent time with such damage in Italian context before the recent Emilia-Romagna Earthquake (May 2012). The main lateral spreading and vertical deformations affected the stability of many buildings and impacted social life inducing valuable lessons on liquefaction risk assessment and remediation. This paper aims first of all to reproduce soil response to liquefaction-induced lateral effects and thus to evaluate stone column mitigation technique effectiveness by gradually increasing the extension of remediation, in order to achieve a satisfactory lower level of permanent deformations. The study is based on the use of a FE computational interface able to analyse the earthquake-induced three-dimensional pore pressure generation adopting one of the most credited nonlinear theories in order to assess realistically the displacements connected to lateral spreading.

The contribution of geophysical techniques to site characterisation and liquefaction risk assessment: Case study of Nafplion City, Greece

Abstract We present an example of how geophysical methodologies can considerably contribute to seismic and liquefaction risk assessment in an area where urban development is planned. The inspection for possible hidden faults by geophysical methods is particularly critical, since such a possibility could practically hinder the town planning in this area. However, even if no primary threats exist within the area, the response of the foundation soil to various scenarios of historical earthquakes, which have affected the place in the past, must be examined. The geophysical methodologies could also assist this analysis, contributing with the calculation of the amplification of the seismic motion from the bedrock up to the surface. The investigation area of Nafplion, Greece, was suspected to have high liquefaction potential since the foundation soil consists of loose sandy silt with a very shallow aquifer. The implementation of gravity and seismic methods considerably aided the investigation for possible seismic faults. Special emphasis was given to seismic depth migration and particularly to the construction of valid velocity models, in order to precisely calculate the dip characteristics of the structures. Shallow seismic techniques were also applied to provide the near-surface velocity structure, which is a prerequisite for assessing the liquefaction risk. In particular, our case study provides an example of how seismic methods (seismic reflection, seismic refraction, seismic modelling, MASW, multichannel analysis of microtremors and crosshole investigations) when combined with geotechnical borehole testing, enhance the reliability of their output and allow the coverage of wide areas in a cost-effective way in comparison to standard borehole tests. Data and information provided by the application of the geophysical methods were subsequently incorporated in the liquefaction risk assessment at several selected sites within the study area. Factors of safety against liquefaction and liquefaction potential values were computed for three scenario earthquakes that were selected on the basis of the known seismic impact of past earthquakes in the town of Nafplion. We found that liquefaction probability can reach values as high as 80% at some sites depending on the selected earthquake scenario. The formations most prone to liquefaction are detected at depths between 5 and 10 m. This information can be helpful for making risk-based design decision in this region.

Liquefaction Risk Assessment Using Geostatistics to account for Soil Spatial Variability

Liquefaction triggering assessments are often performed for individual locations, providing little information in regard to the expected spatial extent of liquefaction events. The present paper proposes a method to quantify the potential extent of liquefaction by accounting for spatial dependence of soil properties and potential future earthquake shaking. Random-field theory and geostatistics tools are used to model soil properties and earthquake shaking intensity; this approach facilitates incorporation of measurement results obtained at individual locations within the area of interest. An empirical liquefaction triggering criterion is then used to model liquefaction occurrence as a function of the random-field realizations. The framework components are briefly described and an example analysis is performed to illustrate the details of the approach. The area of liquefied soil under a building in Adapazari, Turkey, is considered in the example, conditional upon soil property measurements obtained from nearby standard penetration tests.

Liquefaction Risk Assessment: Evaluation of Three Statistical Models

One approach to liquefaction risk assessment is based on a comparison of the earthquake and soil parameters for the site under study to those known to have either caused or not caused liquefaction at other sites during earlier earthquakes. In this study, three statistical models for liquefaction risk assessment are evaluated and compared. Both linear and nonlinear discriminant analyses are performed on the same set of historical data. It is found that the model with an energy-based formulation for the development of pore water pressure results in the least number of misclassified cases of liquefaction and nonliquefaction. The main conclusion derived from this study is that the use of an integral measure of seismic action, viz., dissipated energy, may result in a better assessment of liquefaction risk.

Liquefaction Risk Assessment: Evaluation of Three Statistical Models

One approach to liquefaction risk assessment is based on a comparison of the earthquake and soil parameters for the site under study to those known to have either caused or not caused liquefaction at other sites during earlier earthquakes. In this study, three statistical models for liquefaction risk assessment are evaluated and compared. Both linear and nonlinear discriminant analyses are performed on the same set of historical data. It is found that the model with an energy-based formulation for the development of pore water pressure results in the least number of misclassified cases of liquefaction and nonliquefaction. The main conclusion derived from this study is that the use of an integral measure of seismic action, viz., dissipated energy, may result in a better assessment of liquefaction risk.