Multi-objective feature selection (MOFS) algorithms for prediction of liquefaction susceptibility of soil based on in si
- PDF / 1,270,800 Bytes
- 23 Pages / 439.37 x 666.142 pts Page_size
- 88 Downloads / 203 Views
Multi‑objective feature selection (MOFS) algorithms for prediction of liquefaction susceptibility of soil based on in situ test methods Sarat Kumar Das1 · Ranajeet Mohanty2 · Madhumita Mohanty1 · Mahasakti Mahamaya2 Received: 1 October 2019 / Accepted: 23 May 2020 © Springer Nature B.V. 2020
Abstract The prediction of liquefaction susceptibility for highly unbalanced database with limited and important input parameters is a crucial issue. The proposed multi-objective feature selection algorithms (MOFS) were applied to highly unbalanced databases of in situ tests: standard penetration test (SPT), cone penetration test (CPT) and shear wave velocity (Vs) test. Two multi-objective algorithms, non-dominated sorting genetic algorithm (NSGA-II) and multi-objective symbiotic organisms search algorithm (MOSOS), were coupled with learning algorithms, artificial neural network (ANN) and multivariate adaptive regression spline (MARS) separately to effectively select the optimal parameters and simultaneously minimize the error. The obtained optimal point has approximately equal accuracy in both liquefiable and non-liquefiable conditions for training and testing. The important inputs found for models based on SPT are: (N1)60, amax and Mw; CPT: qc1, amax and CSR and Vs: Vs1, CSR, amax and Mw. The CPT-based models were found to be the most efficient. Keywords ANN · Feature selection · In situ tests · Liquefaction · MARS · MOSOS · Multiobjective optimization · NSGA-II Abbreviations (N1)60 Corrected SPT value AL Accuracy of the class instance related to liquefaction of soil * Sarat Kumar Das [email protected] Ranajeet Mohanty [email protected] Madhumita Mohanty [email protected] Mahasakti Mahamaya [email protected] 1
Department of Civil Engineering, Indian Institute of Technology (Indian School of Mines), Dhanbad, Jharkhand 826004, India
2
Department of Civil Engineering, National Institute of Technology Rourkela, Odisha 769008, India
13
Vol.:(0123456789)
Natural Hazards
amax Peak horizontal ground acceleration in terms of g ANL Accuracy of the class instance related to non-liquefaction of soil BF Basis function BFi Benefit factor c Normalization on exponent CB Correction for borehole diameter CE Correction for hammer efficiency CN Correction for overburden stress CR Correction for short rod length CS Correction for non-standardized sampler configuration CSR Uniform cyclic stress ratio d Critical depth dw Depth of water table e Distribution of error f (X) True function f(x) A function which is approximated by BFs FC Fines content ith front Fi FCI Fines content index FN Number of liquefaction instances misclassified FP Number of non-liquefaction instances misclassified Gmean Geometric mean of the individual accuracies of each class instance Gmean in terms of error rate Gmean(error) L Liquefaction LI Liquefaction index m Number of objective functions, an integer ≥ 2 Mutual Vector Mutual connection between the organisms Xi and Xj Mw Moment magnitude of the eart
Data Loading...
