Evaluation of soil liquefaction potential using ensemble classifier based on grey wolves optimizer (GWO)

被引:1
|
作者
Reddy, Nerusupalli Dinesh Kumar [1 ]
Diksha
Gupta, Ashok Kumar [1 ]
Sahu, Anil Kumar [1 ]
机构
[1] Delhi Technol Univ, Dept Civil Engn, Delhi 110042, India
关键词
Soil liquefaction; Standard penetration test; Long short-term memory; Support vector machines; Grey wolf optimizer; ARTIFICIAL NEURAL-NETWORK; SUPPORT VECTOR MACHINES; ENERGY CONCEPT; SUSCEPTIBILITY; PREDICTION; RESISTANCE; REGRESSION; MODELS;
D O I
10.1016/j.soildyn.2024.108750
中图分类号
P5 [地质学];
学科分类号
0709 ; 081803 ;
摘要
Soil liquefaction is a primary factor in causing destruction during earthquakes. For over four decades, there has been significant improvement in detecting soil liquefaction. At initially, this process was primarily in sandy, clean soils. Because the cost of soil transformation is often significant, exact estimate of liquefaction potential, together with security considerations, might reduce the scheme's fiscal cost. This research proposes to provide a novel soil liquefaction prediction model with three primary stages. It includes, data visualization was performed using correlation matrix and pair plots to determine the dependency and independency of each variable, as well as the entropy of the data to determine the complexity of the data, before deploying a novel liquefaction methodology that included an ensemble model of sophisticated deep learning classifiers of Long short-term memory (LSTM) + Support Vector Machines (SVM) to reduce the reproducibility problem. Improved Correlation characteristics have been used to pick the most essential variables while removing duplicate and unnecessary characteristics. K-fold validation is used to prevent overfitting, a situation when a model is excessively trained on training data but underperforms on new, untested data. Finally, the grey wolf optimizer was used to improve the operation's local minimum values and convergence.
引用
收藏
页数:15
相关论文
共 50 条
  • [31] Integrate Classifier Diversity Evaluation to Feature Selection Based Classifier Ensemble Reduction
    Yao, Gang
    Chao, Fei
    Zeng, Hualin
    Shi, Minghui
    Jiang, Min
    Zhou, Changle
    2014 14TH UK WORKSHOP ON COMPUTATIONAL INTELLIGENCE (UKCI), 2014, : 37 - 43
  • [32] Improved prediction of soil liquefaction susceptibility using ensemble learning algorithms
    Tiwari, Satyam
    Das, Sarat K.
    Mohanty, Madhumita
    Prakhar
    GEOMECHANICS AND ENGINEERING, 2024, 37 (05) : 475 - 498
  • [33] Energy-Based Estimation of Soil Liquefaction Potential Using GMDH Algorithm
    Javdanian, Hamed
    Heidari, Ali
    Kamgar, Reza
    IRANIAN JOURNAL OF SCIENCE AND TECHNOLOGY-TRANSACTIONS OF CIVIL ENGINEERING, 2017, 41 (03) : 283 - 295
  • [34] Energy-Based Estimation of Soil Liquefaction Potential Using GMDH Algorithm
    Hamed Javdanian
    Ali Heidari
    Reza Kamgar
    Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2017, 41 : 283 - 295
  • [35] Evaluation of liquefaction potential based on CPT data using random forest
    V. R. Kohestani
    M. Hassanlourad
    A. Ardakani
    Natural Hazards, 2015, 79 : 1079 - 1089
  • [36] Evaluation of liquefaction potential based on CPT data using random forest
    Kohestani, V. R.
    Hassanlourad, M.
    Ardakani, A.
    NATURAL HAZARDS, 2015, 79 (02) : 1079 - 1089
  • [37] Evaluation of soil liquefaction potential index based on SPT data in the Erzincan, Eastern Turkey
    Duman, E. Subasi
    Ikizler, S. B.
    Angin, Z.
    ARABIAN JOURNAL OF GEOSCIENCES, 2015, 8 (07) : 5269 - 5283
  • [38] Evaluation of soil liquefaction potential index based on SPT data in the Erzincan, Eastern Turkey
    E. Subaşı Duman
    S. B. Ikizler
    Z. Angin
    Arabian Journal of Geosciences, 2015, 8 : 5269 - 5283
  • [39] Evaluation of the impact of environmental conditions on diabetes using ensemble classifier based on genetic algorithm
    Khademi, F.
    Motameni, H.
    Rabbani, M.
    Akbari, E.
    SCIENTIA IRANICA, 2022, 29 (05) : 2394 - 2404
  • [40] Assessment of soil liquefaction potential using MASW method
    Lin, CP
    PROGRESS IN ENVIRONMENTAL AND ENGINEERING GEOPHYSICS, 2004, : 197 - 203