Prediction of efficiency of the filled-trench in layered soil through artificial neural network

Published: Jan. 16, 2025

Purpose Vibrations are transmitted through the ground surface to building foundations, causing distress structures and their occupants. Installing a wave barrier between vibration source buildings is suitable method mitigate vibration. However, complexity arises in selecting right trench design due various influencing parameters. This paper aims present novel predict efficiency of geofoam-filled mitigating vibrations within layered soil using an artificial neural network (ANN). Design/methodology/approach study extends parametric investigation conducted by Naghizadeh (Naghizadehrokni, 2022), where they identified key parameters trench’s efficiency. A multilayered feedforward back-propagation training was developed for prediction task. The ANN model comprises input variables, including location, depth, width trench, thickness shear velocity first layer as well geofoam type. With total 18,750 data points from study, trained validated. Findings accuracy evaluated separate training, validation testing sets. Different configurations were comparing coefficient determination ( R2 ) mean square error. optimal architecture used previous results, revealing effectiveness approach. Furthermore, ANN’s predictive performance compared with finite element results. results indicate high level accuracy, regression R-value 0.98 analysis entire set. Originality/value After studying research, author need evaluate trenches. To meet this requirement, collected 2022) precisely these

Language: Английский

Interface Shear Behavior of Geofoam-Sand: A Comprehensive Laboratory Investigation

International Journal of Geosynthetics and Ground Engineering, Journal Year: 2024, Volume and Issue: 10(6)

Published: Oct. 16, 2024

Language: Английский