Location: Watershed Physical Processes ResearchTitle: Best of SAGEEP: Application of artificial neural network to forecast geotechnical parameters and seismic wave velocity
|JOHORA, F - University Of Mississippi|
|HICKEY, CRAIG - University Of Mississippi|
|YASARER, H - University Of Mississippi|
Submitted to: Meeting Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 8/5/2021
Publication Date: 8/5/2021
Citation: Johora, F.T., Hickey, C.J., Yasarer, H. 2021. Best of SAGEEP: Application of artificial neural network to forecast geotechnical parameters and seismic wave velocity. European Association of Geoscientists and Engineers/ Near Surface Geoscience (EAGE/NSG) conference held August 29 - September 2, 2021 - Bordeaux - Online. 1 p.
Interpretive Summary: Abstract Only
Technical Abstract: Non-destructive geophysical seismic methods are effective for investigating soils without affecting the inherent mechanical properties. The current research is focused on developing models to forecast seismic wave velocity and geotechnical parameters using the artificial neural network (ANN) technique. Published seismic wave velocity, liquid limit, plastic limit, water content, and dry density from in field and laboratory measurements are used to develop ANN models. Several ANN models are developed using data from the field and lab separately and jointly. Models are developed to predict seismic wave velocity using geotechnical parameters, predict water content using seismic wave (plus additional geotechnical parameters) and predicting dry density using seismic wave (plus additional geotechnical parameters) Performance of the ANN models are assessed based on mean absolute relative error (MARE), average squared error (ASE) and coefficient of determination (R2). Multilinear regression analysis is also performed. The results indicate that ANN performs better than multilinear regression analysis.