Prediction of shear wave velocity in underground layers using Particle Swarm Optimization

11th International Conference on Geotechnical Engineering in Tropical Regions;1st International Conference on Highway and Transportation Engineering 2019

矿业工程;运输工程

Anak Upom, Mark Ruben^1 ; Asmawisham Alel, Mohd Nur^1 ; Ab Kadir, Mariyana Aida^1 ; Yuzir, Ali^2

School of Civil Engineering, Universiti Teknologi Malaysia, Johor, Johor Bahru

81310, Malaysia^1

Malaysia-Japan International Institute of Technology (MJIIT), Universiti Teknologi Malaysia, Kuala Lumpur

54100, Malaysia^2

关键词: Coefficient of determination;    Empirical correlations;    Independent parameters;    K nearest neighbor (KNN);    Mean absolute percentage error;    Multi-linear regression;    Penetration resistances;    RMSE (root mean square error);   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/527/1/012012/pdf
DOI  :  10.1088/1757-899X/527/1/012012

来源: IOP

【 摘 要 】

Shear wave velocity (Vs) is considered a key soil parameter in the field of earthquake engineering. The time-averaged shear wave velocity in the upper 30 m (Vs30) layer of soil is used to classify seismic site class. In-situ Vs test is sometimes unsuitable to the project's need due to financial reasons, noisy environment on site or simply the lack of expertise. This paper attempts to develop a global prediction model for Vs using Standard Penetration Resistance (Nspt), depth (z) and soil type (s t) as the independent parameters. Two approaches to modelling would be taken; a multi-linear regression (MLR) model and an ensemble (EN-PSO) model. The EN-PSO model attempts to improve upon the accuracy of the MLR model prediction ability using the ensemble learning method. A dataset was compiled from literatures for this paper. 5 Base models were developed: MLR, Random Forest (RFR), Support Vector Machine (SVR), Artificial Neural Network (ANN) and k-Nearest Neighbor (KNN) which are combined into an ensemble model named EN-PSO. The weights for EN-SPO was then calculated using Particle Swarm Optimization (PSO). The performance of each models were then compared and it was shown that EN-PSO was the best in terms of: MAE (Mean Absolute Error) = 22.085, MAPE (Mean Absolute Percentage Error) = 9.1 %, RMSE (Root Mean Square Error) = 31.741 and R2 Coefficient of Determination) = 0.895. In addition, it was also shown that the EN-PSO model was able to improve upon the performance of the MLR model, which the most accurate among the Base models. Comparisons were also made between EN-PSO and other suggested Universal Vs correlations and EN-PSO was shown to outperform the other correlation based on prediction using a modified Test set. Three new empirical correlations as alternative for the EN-PSO model was also presented.