A Metaheuristic-Trained Wavelet Neural Network for Predicting of Soil Seismic Liquefaction Based on the Standard Penetration Test Results

Document Type : Regular Article

Authors

1 Associate Professor, Department of Civil Engineering, Sirjan University of Technology, Sirjan, Iran

2 Department of Civil Engineering Sirjan University of Technology, Sirjan, Iran

10.22115/scce.2023.408358.1698

Abstract

Earthquake-induced liquefaction is one of the natural hazards that can cause irreparable damages to buildings and infrastructure in addition to threatening human lives. Therefore, accurate prediction of liquefaction has always been one of the important tasks of geotechnical engineers. In the present study, a binary classification model based on the hybrid method of Wavelet Neural Network (WNN) and Teaching-Learning Based Optimization (TLBO) is developed to predict the occurrence of liquefaction based on the Standard Penetration Test (SPT) results. To achieve this goal, a reliable dataset consisting of 288 data points has been used. The optimal architecture of the WNN was defined as 2-8-1, which shows that the optimal WNN has two neurons in the input layer, 8 neurons in the hidden layer with Shannon wavelet function and one neuron in the output layer with Hardlim activation function. The developed WNN model is able to predict liquefaction with an overall accuracy of 96.52% based on two input variables including modified cyclic stress ratio (CSR7.5) and SPT number (N1,60). Comparing the accuracy of the WNN-TLBO model developed in this research with that of other artificial intelligence models (e.g., artificial neural network, multi-gene genetic programming model, relevance vector machine, and support vector machine) developed in this field shows the superior accuracy of the WNN-TLBO model compared to previous models.

Keywords

Main Subjects

Files

History

  • Receive Date: 05 August 2023
  • Revise Date: 06 October 2023
  • Accept Date: 20 November 2023

Share

How to cite

Statistics