An evolutionary based approach for assessment of earthquake-induced soil liquefaction and lateral displacement
Introduction
During an earthquake, significant damage can result due to instability of the soil in the area affected by internal seismic waves. Liquefaction is known as one of the major causes of ground failure due to earthquake. Assessment of liquefaction potential and determination of liquefaction induced lateral spreading are complex geotechnical engineering problems and have attracted considerable attention of geotechnical researchers in the past three decades (Javadi et al., 2006). A number of analytical models have been developed to predict and evaluate liquefaction potential and liquefaction-induced lateral spreading (Youd et al., 2001). In recent years, by pervasive developments in computational software and hardware, several alternative computer aided pattern recognition approaches have emerged. The main idea behind pattern recognition systems such as neural network, fuzzy logic or genetic programming is that they learn adaptively from experience and extract various discriminants, each appropriate for its purpose. Artificial neural networks (ANNs) and multi-layer regression (MLR) are the most widely used pattern recognition procedures that have been introduced for determination of liquefaction occurrence and liquefaction induced lateral displacement (Juang and Chen, 1999, Bartlet and Youd, 1992).
The ANNs models have the ability to operate on large quantities of data and learn complex model functions from examples, i.e., by training on sets of input and output data. The greatest advantage of ANNs over traditional modeling techniques is their ability to capture nonlinear and complex interactions between the variables of a system without having to assume the form of relationship between input and output parameters. In the context of determination of liquefaction occurrence, ANNs can be trained to learn the complex relationship between the soil and earthquake characteristics with the liquefaction potential and lateral spreading, requiring no prior knowledge of the form of the relationship. However, the main disadvantage of the neural network-based approach is the large complexity of the network structure, as it represents the knowledge in terms of a weight matrix together with biases which are not accessible to user (Rezania et al., 2010).
In this paper a new approach is introduced based on evolutionary polynomial regression (EPR) for assessment of the liquefaction potential and lateral spreading due to earthquakes. The EPR models for evaluation of liquefaction potential and lateral spreading are trained and tested separately using two different datasets of field case histories. The data for evaluation of liquefaction potential are collected from different sources in the literature and include 420 field cases. The lateral spreading dataset includes 484 case histories from 11 earthquakes gathered from the literature. The results of the developed models for evaluation of liquefaction and lateral spreading are compared with those obtained from the most commonly used technique in each case. It is shown that the EPR models are able to learn, with a very high accuracy, the complex relationship between liquefaction and its contributing factors in the form of a function; and generalize the learning to provide predictions for new cases that are not used in the construction of the model.
Section snippets
Evolutionary polynomial regression
Evolutionary polynomial regression (EPR) is a data-driven method based on evolutionary computing, aimed to search for polynomial structures representing a system. A general EPR expression can be presented as (Giustolisi and Savic, 2006)where y is the estimated vector of output of the process; aj is a constant; F is a function constructed by the process; X is the matrix of input variables; f is a function defined by the user; and n is the number of terms of the target
Evaluation of liquefaction potential using EPR
It is believed that liquefaction occurs when the pore pressure approaches the confining pressure in loose, saturated sands under earthquake loading where, as a result of a rapid and dramatic loss of soil strength, it can initiate movement of large blocks of soil causing extensive damage to civil engineering structures (Youd et al., 2001). Liquefaction occurrence depends on the mechanical characteristics of the soil layers in the site, the depth of the water table, the intensity and duration of
Evaluation of liquefaction induced lateral displacements using EPR
Liquefaction is known as one of the major causes of ground movement and failure related to earthquake. It can initiate movement of large blocks of soil with amplitudes ranging from a few centimeters to 10 m or more, eventually causing extensive damage to buried utilities, lifeline networks and many other underground and surface civil engineering structures. Liquefaction induced lateral displacement generally occurs on gentle slopes ranging from 0.3° to 3° (Committee on Earthquake Engineering
Summary and conclusion
Assessment of liquefaction potential and determination of liquefaction induced lateral spreading are complex geotechnical engineering problems. In this paper a new approach was presented, based on evolutionary polynomial regression, for assessment of the liquefaction potential and lateral spreading due to earthquakes. Separate EPR models were developed and tested for evaluation of liquefaction potential and lateral spreading using two different datasets of actual field case histories.
A robust
References (36)
- et al.
Evaluation of lateral spreading using artificial neural networks
Soil Dynamics and Earthquake Engineering
(2005) - et al.
Evaluation of liquefaction induced lateral displacements using genetic programming
Computers and Geotechnics
(2006) - et al.
Evaluation of liquefaction potential based on CPT results using evolutionary polynomial regression
Computers and Geotechnics
(2010) - et al.
Evaluation of liquefaction potentials of soils using cone penetration tests
Soils and Foundations, Tokyo, Japan
(1988) - Bartlet, S.F., Youd, T.L., 1992. Empirical analysis of horizontal ground displacement generated by liquefaction-induced...
- et al.
Empirical prediction of liquefaction-induced lateral spread
Journal of the Geotechnical Engineering Division
(1995) Liquefaction of Soils during Earthquakes
(1985)- et al.
An investigation on stream temperature analysis based on evolutionary computing
Hydrological Processes
(2008) - et al.
A symbolic data-driven technique based on evolutionary polynomial regression
Journal of Hydroinformatics
(2006) - et al.
Advances in data-driven analyses and modelling using EPR-MOGA
Journal of Hydroinformatics
(2009)
An evolutionary multi-objective strategy for the effective management of groundwater resources
Water Resources Research
Seismic liquefaction potential assessed by neural networks
Journal of the Geotechnical Engineering, ASCE
Neural network modeling of CPT seismic liquefaction data
Journal of the Geotechnical Engineering, ASCE
Evaluation of liquefaction potential of soil deposits using artificial neural networks
Engineering Computations
CPT-based liquefaction evaluation using artificial neural networks
Computer-Aided Civil and Infrastructure Engineering
CPT-based liquefaction analysis, Part1: determination of limit state function
Géotechnique
Simplified cone penetration test-based method for evaluating liquefaction resistance of soils
Journal of the Geotechnical and Geoenvironmental Engineering, ASCE
Cited by (76)
Smart prediction of liquefaction-induced lateral spreading
2023, Journal of Rock Mechanics and Geotechnical EngineeringA hybrid multi-step sensitivity-driven evolutionary polynomial regression enables robust model structure selection
2022, Engineering Applications of Artificial IntelligenceBayesian hierarchical and measurement uncertainty model building for liquefaction triggering assessment
2021, Computers and GeotechnicsCitation Excerpt :This is referred to as over-fitting in the predictive modeling world (Kuhn and Johnson, 2013). To date, only a few relevant models (Oommen et al., 2010, Rezania et al., 2011, etc.) split their databases into training and testing sets through cross validation or other methods to develop relatively unbiased metrics of model performance. Finally, existing models do not account for intra-event correlation between outcomes or sample size discrepancies between events.
Soil Liquefaction Assessment by CPT and V<inf>S</inf> Data and Incomplete-Fuzzy C-Means Clustering
2024, Geotechnical and Geological EngineeringIntegration of Double-Weighted Bayesian and Simplified Methods for Predicting Seismic Liquefaction Based on Multiple Databases
2023, International Journal of GeomechanicsWhy “AI” models for predicting soil liquefaction have been ignored, plus some that shouldn’t be
2023, Earthquake Spectra