Artificial neural network and optimization algorithm to improve soil resistance by means of aggregation size variation

Nowadays, extensive researches are conducted on soil improvement and most of them are focused on the enhancement of soil strength properties. In this study, the aggregation of soil particles including Gravel, Sand, and Clay was investigated on soil resistance. To do this, the number of 25 different testing experiments were experimentally performed according to the Design of Experiment (DOE). The soil strength parameter was determined by means of direct shear tests on large samples. Based on the experimental results, the effect of aggregation size on shear strength was extensively investigated. To do this, Artificial Neural Network (ANN) was employed to evaluate the effect of different combinations of soil particles on bearing capacity. To provide more precise predictive model, the ANN was trained using the Genetic Algorithm (GA). After that, the optimal condition of mix design with optimal load-bearing capacity was found by a hybrid strategy of EOA-ANN. It was obtained 1091.94 (KPa) that had not been reported using the experiments. To evaluate the reliability of the proposed method, a new experimental test was conducted based on the optimal parameters found by hybrid strategy and it was verified with corresponding estimated ones. At the end, very good agreement was reported between the new experiment and the optimal results. It shows that the employed intelligent method in this paper can be successfully used for the enhancement of soil properties in other investigations. Review History: Received: 16 September 2018 Revised: 31 October 2018 Accepted: 12 December 2018 Available Online: 12 December 2018

[1]  M. Coop,et al.  The mechanics of an Italian silt: an example of ‘transitional’ behaviour , 2006 .

[2]  Amin Barari,et al.  Determination of the ultimate limit states of shallow foundations using gene expression programming (GEP) approach , 2015 .

[3]  Zheng-Yi Feng,et al.  Dynamic Properties of Granulated Rubber/Sand Mixtures , 2000 .

[4]  Prabir Kumar Basudhar,et al.  Prediction of residual friction angle of clays using artificial neural network , 2008 .

[5]  Yacoub M. Najjar,et al.  Simulating the stress–strain behavior of Georgia kaolin via recurrent neuronet approach , 2007 .

[6]  L. Zhuang,et al.  Influence of fines content on the undrained cyclic shear strength of sand–clay mixtures , 2016 .

[7]  Pijush Samui,et al.  Application of Artificial Intelligence to Maximum Dry Density and Unconfined Compressive Strength of Cement Stabilized Soil , 2011 .

[8]  J. C. Santamarina,et al.  Sand–rubber mixtures (large rubber chips) , 2008 .

[9]  Sibel Pamukcu,et al.  Thermoelastic Enhancement of Damping of Sand Using Synthetic Ground Rubber , 2006 .

[10]  V. T. Phan,et al.  Engineering behavior and correlated parameters from obtained results of sand–silt mixtures , 2015 .

[11]  Vishal Saxena,et al.  A Thesis Presented to The Academic Faculty , 2003 .

[12]  O. Carașca.  Soil Improvement by Mixing: Techniques and Performances , 2016 .

[13]  Nader Nariman-zadeh,et al.  Piles shaft capacity from CPT and CPTu data by polynomial neural networks and genetic algorithms , 2009 .

[14]  Young-Su Kim,et al.  Liquefaction Resistance of Sands Containing Plastic Fines with Different Plasticity , 2013 .

[15]  K. Terzaghi.  Theoretical Soil Mechanics , 1943 .

[16]  Ali Firat Cabalar,et al.  Fall cone tests on clay–sand mixtures , 2015 .

[17]  Lars Bo Ibsen,et al.  Undrained response of bucket foundations to moment loading , 2012 .

[18]  Domenico Umbrello,et al.  Experimental Investigation to Optimize Tool Life and Surface Roughness in Inconel 718 Machining , 2016 .

[19]  Binh Thai Pham,et al.  Prediction of shear strength of soft soil using machine learning methods , 2018, CATENA.

[20]  Aminul Islam Laskar,et al.  Prediction of unconfined compressive strength of geopolymer stabilized clayey soil using Artificial Neural Network , 2015 .

[21]  Hyunki Kim,et al.  Spatial variability in soils: stiffness and strength , 2005 .

[22]  D. P. Kanungo,et al.  Artificial Neural Network (ANN) and Regression Tree (CART) applications for the indirect estimation of unsaturated soil shear strength parameters , 2014, Frontiers of Earth Science.

[23]  Prabir Kumar Basudhar,et al.  Undrained lateral load capacity of piles in clay using artificial neural network , 2006 .

[24]  Hossein Amirabadi,et al.  Experimental measurement and optimization of tensile residual stress in turning process of Inconel718 superalloy , 2015 .

[25]  Hossein Amirabadi,et al.  Application of artificial neural network and optimization algorithms for optimizing surface roughness, tool life and cutting forces in turning operation , 2013 .

[26]  Hossein Amirabadi,et al.  Simultaneous Optimizing Residual Stress and Surface Roughness in Turning of Inconel718 Superalloy , 2014 .

[27]  S. Kiran,et al.  Shear Strength Prediction of Soil based on Probabilistic Neural Network , 2016 .

[28]  Hamza Güllü,et al.  Prediction of peak ground acceleration by genetic expression programming and regression: A comparison using likelihood-based measure , 2012 .

[29]  Matthew Richard Coop,et al.  Some aspects of the compressibility behaviour of a clayey sand , 2001 .