Geomechanics and Engineering
Volume 45, Number 4
DOI: 485-516
Evaluation and optimization of effective depth of dynamic compaction using sugeno fuzzy inference system and PSO
Hamid Koohsari , Hamid Alielahi , Amir Najafi , Mohammad Adampira
Abstract
Dynamic Compaction (DC) is a widely adopted ground improvement technique, particularly effective for granular soils. This study presents a hybrid approach combining fuzzy logic implemented via a Sugeno Inference System and Particle Swarm Optimization (PSO) to estimate and enhance the effective depth of DC. A fuzzy model was developed to evaluate the influence of key operational parameters, including tamper weight, drop height, radius, number of drops, grid spacing, and soil resistance. A symbolic regression-based correlation was proposed to estimate the improvement depth and was validated against the fuzzy model results. Parametric analysis revealed that the interaction between tamper weight and drop height is the most influential factor. Optimization using PSO resulted in a 33% increase in the maximum improvement depth without additional energy input. In addition, optimal design parameters were identified, including a tamper radius of 1.5–2.0 m, 25 drops per point, and a grid spacing of 6–7 m. This hybrid AI-based framework offers a practical alternative to conventional empirical DC design and demonstrates promising capability for improving design efficiency and parameter selection.
Key Words
dynamic compaction (DC); estimated improvement depth; fuzzy system; granular soils; particle swarm optimization (PSO)
Address
- Hamid Koohsari, Hamid Alielahi — Department of Civil Engineering, Za.C., Islamic Azad University, Zanjan, Iran
- Amir Najafi — Department of Industrial Engineering, NT.C, Islamic Azad University, Tehran, Iran
- Mohammad Adampira — Department of Civil Engineering, SR.C., Islamic Azad University, Tehran, Iran
PDF Viewer
Preview is limited to the first 3 pages. Sign in to access the full PDF.
Loading…
