| Abstrakt: |
Dynamic compaction (DC) with tamping is one of the soil-improvement methods. It has been widely accepted due to its advantages over other soil-improvement methods. In this paper, an attempt has been made to use fuzzy logic and the Sugeno inference system to investigate the effect of the parameters involved in soil-improvement operations using the DC method on the improvement depths of relative degrees. The input variables used for loose granular soils include the tamper weight, the height of tamping, the tamper radius, the grid spacing, the number of drops and the soil layer geotechnical properties. Four improvement depths of relative degrees are used as the output variables. In addition, a correlation has been proposed to estimate the effective depth of granular soils and to make a comparison to the fuzzy results. In order to achieve the maximum improvement depth, the Particle Swarm Optimization (PSO) algorithm has been used. The results indicate that the most efficient input parameters are the tamper weight, the height of tamping and the grid spacing. Moreover, the interaction between the tamper weight and the height of tamping plays the most important role in the design procedure. The results show that when using PSO, the maximum depth of improvement is increased by 33%. Studies reveal that the optimal tamper radius for most compaction patterns with medium to high applied energies is 1.5–2m, the optimal number of drops is 25 and the optimal grid spacing is 6–7m. It should be noted that the maximum improvement depth has been achieved with the PSO dynamic compaction pattern. |
