Comparing Settlements of Foundations Improved with Stone Column and Pile Using PLAXIS and Artificial Neural Network (Case Study of Mokran Region)

Comparing Settlements of Foundations Improved with Stone Column and Pile Using PLAXIS and Artificial Neural Network (Case Study of Mokran Region)

Mohammad Arabameri1 Hamed Zarei2 Somaye Akbarnezhad3

1) Ph.D Geotechnical Student, Amirkabir University of
2) Master’s Degree in the field of Geotechnical
3) Master’s Degree in the field of Geotechnical Engineering, Technical experts and Soil improvement at Shaloodeh Khak Consulting Engineers

IOI Code : XYBD-ZKFDF
Publication : 4th International Conference On Research Science And Technology(4rstconf.com)
Abstract :
    With the increasing growth of population, development of industries, and shortage of lands suiting construction of required structures, human has showed tendency for using lands with lower engineering quality. Soil improvement with stone columns is one of the most suitable methods for improvement of soft and weak soils due to its environment-friendliness, cost-effectiveness, decreased settlement, loadbearing capacity, etc. This technique can be applied to clay and silt soil as well as sand soil. The study area in this research was located in Chabahar Province, one of the major problems of which is the loose soil in need of improvement. In this research, to determine the region’s soil properties and soil resistive parameters first different boreholes (to different depths) were dug in the region, and then the stone column and pile improvement (retrofitting) methods were used for modeling in Plaxis discrete element software. Afterwards, to assess validity of results a network of different data types was trained in different conditions and was tested within the research scope. Finally, the results were compared to Plaxis results. It was found that settlement prediction of the foundation retrofitted with the artificial neural network (ANN) not only was time- and cost-effective, but also displayed more efficiency than the other method under study. In the end, the stone column method was selected as the optimal method considering technical and economic dimensions.

Keywords : settlement bearing capacity stone column pile artificial neural network plaxis