Dimensionality reduction and prediction of soil consolidation coefficient using random forest coupling with Relief algorithm

The consolidation coefficient of soil(C_(v))is a crucial parameter used for the design of structures leaned on soft soi.In general,the C_(v) is determined experimentally in the laboratory.However,the experimental tests are time-consuming as well as expensive.Therefore,researchers tried several ways to determine C_(v) via other simple soil parameters.In this study,we developed a hybrid model of Random Forest coupling with a Relief algorithm(RF-RL)to predict the C_(v) of soil.To conduct this study,a database of soil parameters collected from a case study region in Vietnam was used for modeling.The performance of the proposed models was assessed via statistical indicators,namely Coefficient of determination(R^(2)),Root Mean Squared Error(RMSE),and Mean Absolute Error(MAE).The proposal models were constructed with four sets of soil variables,including 6,7,8,and 13 inputs.The results revealed that all models performed well with a high performance(R^(2)>0.980).Although the RF-RL model with 13 variables has the highest prediction accuracy(R^(2)=0.9869),the difference compared with other models was negligible(i.e.,R^(2)=0.9824,0.9850,0.9825 for the cases with 6,7,8 inputs,respectively).Thus,it can be concluded that the hybrid model of RF-RL can be employed to predict C_(v) based on the basic soil parameters.

CONSOLIDATION THEORY OF UNSATURATED SOIL BASED ON THE THEORY OF MIXTURE(Ⅱ)

The present paper uses the mathematics model for consolidation of unsaturated soil developed in ref. [I] to solve boundary value problems. The analytical solutions for one-dimensional consolidation problem are gained by making use of Laplace transform and finite Fourier transform. The displacement and the pore water pressure as well as the pore gas pressure are found from governing equations simultaneously. The theoretical formulae of coefficient and degree of consolidation are also given in the paper. With the help of the method of Galerkin Weighted Residuals, the finite element equations for two-dimensional consolidation problem are derived. A FORTRAN program named CSU8 using 8-node isoparameter element is designed. A plane strain consolidation problem is solved using the program, and some distinguishing features on consolidation of unsaturated soil and certain peculiarities on numerical analysis are revealed. These achievements make it convenient to apply the theory proposed by the author in engineering practice.

Some Classical Terzaghi Primary Consolidation Differential Equation Mathematical Added Variabilities in Order to Get Better Field Monitoring Data Match Results

In this paper is illustrated a mathematical added variability of the classical Terzaghi primary consolidation equation(1923)considering independently as variables the Consolidation Coefficient Cv and as well the Height Hi of the consolidating Laboratory Consolidation Test soil sample in order to finally grasp the low permeability layer time behaviour.It is easy to show that,when the Cv variation is positive,each of these two added variabilities differentiations has as maximum a factor 2 related to the laboratory evaluated coefficient of consolidation,for a certain incremental load of reference in a Laboratory Consolidation Test.At this scope,it is analysed the overall behaviour of a typical clayey material,from the mineralogical point of view,namely especially either composed by lean clay with main kaolinite mineralogical content or fat namely with Illite mineralogical content or even very dilatant namely principally constituted by Montmorillonite.The Montmorillonite variability with Cv is negative,and consequently the differentiation enhancement factor can become naught.As it is known so far,in normal conditions of a soft clay,a difference in Construction Values of the Coefficient of Consolidation is up to 23 times greater than laboratory evaluated results,and this according to the author’s experience,may be also mainly explained not starting from Laboratory Consolidation Test Data,but through a more general macroscopic behaviour of the soil underneath the newly loaded area,putting aside the case of temperature-induced changes.In conclusion,it is suggested how to model the analytical problem of the so modified Terzaghi Primary Consolidation differential equation in order to better manage the construction unknowns of the phenomenon.

Delta River Swampy Areas Earth Fill Embankments Primary Consolidation Management Issues

It is well known that soft silty clayey and even peaty soils especially existing in Great River Deltas Swampy Areas,under important Earth Fill Embankment Construction experience huge and hardly bearable primary consolidations settlements along with the minor but not negligible consequent secondary consolidation effects.In order to properly manage these particular huge settlements environments,it is very important to follow up the settlements monitoring data,by a macroscopic soil volume interpretation along with some amendments namely some mathematical added variabilities of the classic Terzaghi Primary Consolidation Equation,which are examined in a companion paper recently published in this Journal.In this paper some indications are given about how to face the macroscopic soil volume primary consolidation settlements,and especially it is suggested how to interpret the misleading laboratory consolidation test values of the coefficient of consolidation.Moreover,some design alternative solutions are examined to grasp both the potential technical and economic benefits along with their consequent disadvantages.Finally,this paper underlined the primary role of the supervision engineer to get a good estimate in the settlements forecasting and his related ability to understand the meaning of anomalous monitoring data and to timely make and match the primary consolidation settlements forecasting calculation adjustments.